Dissertationen: „Transfer of adaptation“

1

Garrick-Bethell, Ian 1980. „Cross plane transfer of vestibular adaptation to human centrifugation“. Thesis, Massachusetts Institute of Technology, 2004. http://hdl.handle.net/1721.1/17770.

Der volle Inhalt der Quelle

Annotation:

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 2004.
Includes bibliographical references (p. 101-106).
Human short-radius centrifugation (SRC) is being investigated as a volume-efficient means of delivering intermittent doses of "artificial gravity" to counter the deleterious effects of long exposures to weightlessness. Rotation rates on short radius centrifuges are high to provide the needed g-loading, and therefore entail a variety of unusual vestibular stimuli when certain head movements are made. Since these movements can elicit inappropriate nystagmus, illusions of tumbling, and motion sickness, efforts have been made to adapt people to the stimuli. So far these efforts have been successful in showing that people will adapt to at least one plane of head motion, the yaw (transverse) plane, during supine head-on-axis rotation. However, astronauts must be adapted to all planes of head motion if they are to function normally on the centrifuge. If adaptation to yaw head turns transferred to some extent to pitch (sagittal) plane turns, or any other plane of motion, it would greatly simplify and hasten the adaptation process. To investigate if transfer of adaptation across planes is possible, 10 subjects in the Experimental Group performed a sufficient number of yaw plane head turns to demonstrate adaptation. Adaptation was indicated by decreases in metrics of the off-axis vestibuloocular reflex induced by the head turns, and by subjective ratings of illusory motion. A block of pitch movements was performed before and after the yaw movements, and these two pitch blocks were compared to assess how much adaptation to pitch head turns had taken place. The same procedure was followed on a subsequent day. A Control Group of 10 subjects performed only the blocks of pitch turns, and their adaptation was compared to the adaptation to pitch turns measured in the Experimental
(cont.) Group. While both Control and Experimental Groups showed statistically significant signs of adaptation to pitch head turns, we failed to find any significant differences between the amounts of adaptation. If true, this result implies that adaptation to SRC may need to be performed one plane of motion at a time. Additionally, it implies that the brain and vestibular system does not build up a generalized model of SRC stimulation, but rather builds adaptation one input at a time.
by Ian Garrick-Bethell.
S.M.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

2

Otte, Ellen. „Transfer of adaptation across movement categories in eye hand coordination /“. kostenfrei, 2008. http://nbn-resolving.de/urn:nbn:de:hbz:294-26534.

Der volle Inhalt der Quelle

APA, Harvard, Vancouver, ISO und andere Zitierweisen

3

Backlund, Per. „Development process knowledge transfer through method adaptation, implementation, and use /“. Kista : Department of Computer and Systems Sciences, Stockholm University, 2004. http://urn.kb.se/resolve?urn=urn:nbn:se:su:diva-287.

Der volle Inhalt der Quelle

APA, Harvard, Vancouver, ISO und andere Zitierweisen

4

McFarlane, B. „Novel into film: Transfer and adaptation; the processes of transposition“. Thesis, University of East Anglia, 1987. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.374692.

Der volle Inhalt der Quelle

APA, Harvard, Vancouver, ISO und andere Zitierweisen

5

Lefumat, Hannah. „Interlimb transfer of sensorimotor adaptation : predictive factors and underlying processes“. Thesis, Aix-Marseille, 2016. http://www.theses.fr/2016AIXM4014/document.

Der volle Inhalt der Quelle

Annotation:

L’adaptation motrice renvoie à la capacité de notre système nerveux à produire continuellement des mouvements précis et ce malgré le fait que notre environnement ainsi que notre corps puissent être soumis à des modifications. Le transfert d’adaptation entre les membres découle de notre habilité à généraliser ce que l’on a appris, par exemple, avec un bras au bras opposé. Le transfert entre les membres est un objet d’étude complexe. Les conditions amenant au transfert sont largement débattues dans la littérature car les résultats d’une étude à l’autre peuvent être contradictoires. Ce travail de thèse s’inscrit dans une tentative d’apporter une explication concernant l’hétérogénéité des performances et les divergences observées dans les différentes études portant sur le transfert entre les membres. Les deux premières expériences avaient pour but d’identifier si des conditions paradigmatiques ou idiosyncratiques pouvaient influencer les performances du transfert au bras opposé. L’objectif de la troisième expérience était d’étudier l’influence des processus sous-jacents à l’adaptation sur le transfert entre les membres d’après le modèle de Smith et collaborateurs (2006). Nos résultats nous ont permis d’éclaircir certains aspects du transfert concernant les facteurs prédictifs et les processus mis en jeu. Nos deux premières études suggèrent que les différences individuelles sont une source d’information pertinente pour expliquer certains comportements tels que le transfert entre les membres. Notre troisième étude nous a permis de caractériser les processus qui, durant l’adaptation, prédisposent au transfert
Motor adaptation refers to the capacity of our nervous system to produce accurate movements while the properties of our body and our environment continuously change. Interlimb transfer is a process that directly stems from motor adaptation. It occurs when knowledge gained through training with one arm change the performance of the opposite arm movements. Interlimb transfer of adaptation is an intricate process. Numerous studies have investigated the patterns of transfer and conflicted results have been found. The attempt of my PhD project was to identify which factors and processes favor interlimb transfer of adaptation and thence may explain the discrepancies found in the literature. The first two experiments aimed at investigated whether paradigmatic or idiosyncratic features would influence the performance in interlimb transfer. The third experiment provided some insights on the processes allowing interlimb transfer by using the dual-rate model of adaptation put forth by Smith et al. (2006). Our results show that inter-individual differences may be a key factor to consider when studying interlimb transfer of adaptation. Also, the study of the different sub-processes of adaptation seems helpful to understand how interlimb transfer works and how it can be related to other behaviors such as the expression of motor memory

APA, Harvard, Vancouver, ISO und andere Zitierweisen

6

Buechel, Kathryn Jean. „Institutional Adaptation and Public Policy Practices of Military Transfer Credit“. Diss., Virginia Tech, 2020. http://hdl.handle.net/10919/96791.

Der volle Inhalt der Quelle

Annotation:

Veterans who served our country, return with a wealth of experience that transfer into military credit for prior service. These transfer credits in institutions of higher education apply towards education degree attainment. With colleges and universities implementing individual policies for acceptance of credits, veterans experience a loss of credits leading to a duplication of required classes to achieve degrees. To understand inconsistent practices, both federal and institutions of higher education polices are examined. Framed by institutionalization theory, this research sheds light on the public policy process and administration of credit at the organization over time. The study provides findings for how the largest public college and higher education institution in the state of California awards academic credit for military education. Evidence suggests that public higher education institutions adapt based on effective leaders who define and defend the organization's institutional values and mission. This study provides findings on institutional adaptations to create policies and practices that public administrators use to apply transfer military credit into postsecondary academic credit. The focus is on postsecondary credit transferred, or articulated, by entering military first-year students using the GI Bill. The study asks how have major institutions of higher education formalized institutional policies and practices on awarding academic credit for military education?
Doctor of Philosophy
This study provides findings on institutional adaptations to create policies and practices that public administrators use to apply transfer military credit into postsecondary academic credit. The focus is on postsecondary credit transferred, or articulated, by entering military first-year students using the GI Bill. The study asks how have major institutions of higher education formalized institutional policies and practices on awarding academic credit for military education?

APA, Harvard, Vancouver, ISO und andere Zitierweisen

7

Meftah, Sara. „Neural Transfer Learning for Domain Adaptation in Natural Language Processing“. Thesis, université Paris-Saclay, 2021. http://www.theses.fr/2021UPASG021.

Der volle Inhalt der Quelle

Annotation:

Les méthodes d’apprentissage automatique qui reposent sur les Réseaux de Neurones (RNs) ont démontré des performances de prédiction qui s'approchent de plus en plus de la performance humaine dans plusieurs applications du Traitement Automatique de la Langue (TAL) qui bénéficient de la capacité des différentes architectures des RNs à généraliser à partir des régularités apprises à partir d'exemples d'apprentissage. Toutefois, ces modèles sont limités par leur dépendance aux données annotées. En effet, pour être performants, ces modèles neuronaux ont besoin de corpus annotés de taille importante. Par conséquent, uniquement les langues bien dotées peuvent bénéficier directement de l'avancée apportée par les RNs, comme par exemple les formes formelles des langues. Dans le cadre de cette thèse, nous proposons des méthodes d'apprentissage par transfert neuronal pour la construction d'outils de TAL pour les langues peu dotées en exploitant leurs similarités avec des langues bien dotées. Précisément, nous expérimentons nos approches pour le transfert à partir du domaine source des textes formels vers le domaine cible des textes informels (langue utilisée dans les réseaux sociaux). Tout au long de cette thèse nous proposons différentes contributions. Tout d'abord, nous proposons deux approches pour le transfert des connaissances encodées dans les représentations neuronales d'un modèle source, pré-entraîné sur les données annotées du domaine source, vers un modèle cible, adapté par la suite sur quelques exemples annotés du domaine cible. La première méthode transfère des représentations contextuelles pré-entraînées sur le domaine source. Tandis que la deuxième méthode utilise des poids pré-entraînés pour initialiser les paramètres du modèle cible. Ensuite, nous effectuons une série d'analyses pour repérer les limites des méthodes proposées ci-dessus. Nous constatons que, même si l'approche d'apprentissage par transfert proposée améliore les résultats du domaine cible, un transfert négatif « dissimulé » peut atténuer le gain final apporté par l'apprentissage par transfert. De plus, une analyse interprétative du modèle pré-entraîné, montre que les neurones pré-entraînés peuvent être biaisés par ce qu'ils ont appris du domaine source, et donc peuvent avoir des difficultés à apprendre des « patterns » spécifiques au domaine cible. Issu de notre analyse, nous proposons un nouveau schéma d'adaptation qui augmente le modèle cible avec des neurones normalisés, pondérés et initialisés aléatoirement qui permettent une meilleure adaptation au domaine cible tout en conservant les connaissances apprises du domaine source. Enfin, nous proposons une approche d’apprentissage par transfert qui permet de profiter des similarités entre différentes tâches, en plus des connaissances pré-apprises du domaine source
Recent approaches based on end-to-end deep neural networks have revolutionised Natural Language Processing (NLP), achieving remarkable results in several tasks and languages. Nevertheless, these approaches are limited with their "gluttony" in terms of annotated data, since they rely on a supervised training paradigm, i.e. training from scratch on large amounts of annotated data. Therefore, there is a wide gap between NLP technologies capabilities for high-resource languages compared to the long tail of low-resourced languages. Moreover, NLP researchers have focused much of their effort on training NLP models on the news domain, due to the availability of training data. However, many research works have highlighted that models trained on news fail to work efficiently on out-of-domain data, due to their lack of robustness against domain shifts. This thesis presents a study of transfer learning approaches, through which we propose different methods to take benefit from the pre-learned knowledge on the high-resourced domain to enhance the performance of neural NLP models in low-resourced settings. Precisely, we apply our approaches to transfer from the news domain to the social media domain. Indeed, despite the importance of its valuable content for a variety of applications (e.g. public security, health monitoring, or trends highlight), this domain is still poor in terms of annotated data. We present different contributions. First, we propose two methods to transfer the knowledge encoded in the neural representations of a source model pretrained on large labelled datasets from the source domain to the target model, further adapted by a fine-tuning on few annotated examples from the target domain. The first transfers contextualised supervisedly pretrained representations, while the second method transfers pretrained weights, used to initialise the target model's parameters. Second, we perform a series of analysis to spot the limits of the above-mentioned proposed methods. We find that even if the proposed transfer learning approach enhances the performance on social media domain, a hidden negative transfer may mitigate the final gain brought by transfer learning. In addition, an interpretive analysis of the pretrained model, show that pretrained neurons may be biased by what they have learned from the source domain, thus struggle with learning uncommon target-specific patterns. Third, stemming from our analysis, we propose a new adaptation scheme which augments the target model with normalised, weighted and randomly initialised neurons that beget a better adaptation while maintaining the valuable source knowledge. Finally, we propose a model, that in addition to the pre-learned knowledge from the high-resource source-domain, takes advantage of various supervised NLP tasks

APA, Harvard, Vancouver, ISO und andere Zitierweisen

8

Batikh, Ali. „Saccadic adaptation : cross-modal transfer and effect of spatial attention“. Electronic Thesis or Diss., Lyon 1, 2024. http://www.theses.fr/2024LYO10354.

Der volle Inhalt der Quelle

Annotation:

Notre cerveau génère continuellement des mouvements oculaires saccadiques et maintient leur précision grâce à l'adaptation saccadique (AS). Bien que l’AS ait été largement étudiée depuis la fin du 20ᵉ siècle, de nombreuses questions restent ouvertes. Par exemple, en plus des cibles visuelles, les saccades peuvent également être effectuées vers des stimuli somatosensoriels et auditifs, mais on ne sait pas si ces ‘saccades non visuelles’ peuvent être adaptées comme les ‘saccades visuelles’. Dans la première partie de cette thèse, nous avons étudié la possibilité de modifier l'amplitude des saccades vers des cibles tactiles (Étude 1) et auditives (Étude 2) via le paradigme du double saut de cible, largement utilisé pour adapter les saccades visuelles (McLaughlin, 1967). Nous avons également étudié le transfert bidirectionnel et intermodal de l'AS entre les saccades réactives (SR) visuelles et tactiles, ainsi qu'entre les SR visuelles et auditives, respectivement. Nos résultats montrent que les saccades tactiles et auditives peuvent être adaptées d’une manière similaire aux SR visuelles. Cependant, les transferts étaient asymétriques : l'adaptation des SR visuelles transfère complètement aux saccades non adaptées tactiles et auditives, tandis que pour l'adaptation des SR tactiles et auditives, malgré leur généralisation aux saccades non adaptées de la même modalité, le transfert aux SR visuelles n’est que partiel. D'une part, le transfert complet de l'adaptation des SR visuelles soutient l'hypothèse actuelle d'un niveau moteur de l'adaptation de ces saccades. D'autre part, les faibles transferts d'adaptation vers les SR visuelles suggèrent l'existence de sites d'adaptation spécifiques aux SR non visuelles, situés en amont de la voie motrice finale commune à toutes les saccadies. La première partie de cette thèse montre que l'AS peut être déclenchée dans l'obscurité, sans information visuelle sur la position post-saccadique de la cible. Cela semble contredire les théories actuelles sur la nature des signaux d'erreur responsables de l'AS, qui reposent toutes sur ce feedback visuel post-saccadique. Un facteur potentiel qui pourrait servir de signal d'erreur pour l'AS est le locus de l'attention spatiale, comme le suggèrent certaines études précédentes. L'orientation covert (sans mouvement des yeux) de l’attention et les saccades contribuent ensemble à la perception visuelle et impliquent des substrats neuronaux communs. De plus, de récentes études montrent que l'AS module l'orientation de l'attention spatiale, tandis que l'effet inverse, de l'attention spatiale sur l'AS, reste indéterminé. Dans la deuxième partie de cette thèse (Étude 3), nous visons à évaluer en profondeur l’effet modulateur de l'attention spatiale sur l'AS. Nous avons utilisé une combinaison du paradigme du double saut de cible (pour induire l'adaptation) et du paradigme d'orientation intermodale de l’attention pour étudier l'effet de l'attention spatiale exogène et endogène tactile sur l'adaptation des saccades réactives et volontaires, respectivement. Nos résultats montrent des corrélations significatives entre l’adaptation (taux et vitesse) et la quantité d'attention allouée vers ou à l'opposé de la cible saccadique adaptée. Ainsi, l'Étude 3 apporte des arguments supplémentaires en faveur d'un couplage entre l'attention spatiale et l'AS, possiblement via un effet de l'attention spatiale sur les signaux d'erreur saccadique au niveau du cortex pariétal postérieur. Ce travail apporte de nouvelles perspectives sur le contrôle de la précision des SR non visuelles et met en lumière le rôle de l'attention spatiale dans l'AS. Bien que des progrès significatifs aient été réalisés par les modèles étudiant la nature des signaux d'erreur qui conduisent à l'AS, ces derniers ne prennent actuellement pas en compte le couplage entre l'attention spatiale et l'AS, ce qui devrait être le sujet de travaux futurs
Our brain continuously generates saccadic eye movements and maintains their accuracy thanks to saccadic adaptation (SA). Despite this plasticity-based mechanism has been widely studied since the late 20th century, many questions remain unanswered. For instance, in addition to visual targets, saccades can also be performed toward somatosensory and auditory stimuli, but whether these ‘non-visual saccades’ can be subject to similar adaptive mechanisms as visual saccades is unknown. In the first part of this thesis, we investigated the possibility of adapting the amplitude of reactive saccades (RS) to tactile (Study 1) and auditory targets (Study 2) via the double target step paradigm, which has largely been used to induce adaptation of visual saccades since its introduction (McLaughlin 1967). We also investigated the bidirectional cross-modal transfer of adaptation between visual and tactile saccades, as well as between visual and auditory saccades, respectively. Our results revealed that tactile and auditory saccades can be adapted in much the same way as visual saccades. However, the transfer patterns were asymmetric: visual SA transferred fully to non-adapted tactile and auditory saccades, whereas tactile and auditory SA, despite complete generalization to saccades of the same modality but toward non-adapted locations, transferred only partially to the non-adapted visual saccades. On the one hand, the full transfer of visual saccades adaptation further supports the current view of a motor adaptation locus for visual RS. On the other hand, the low adaptation transfers to visual saccades suggest the presence of adaptation loci specific to non-visual RS and situated upstream of the final motor pathway common to all saccades. Interestingly, both studies also demonstrate that SA can be elicited in darkness, thus, without the vision of the post-saccadic target location. This seems to contradict current theories on the nature of error signals driving adaptation, which all rely on post-saccadic visual feedback. One potential factor that might serve as an error signal for SA is the locus of spatial attention, as suggested by some previous studies. Spatial attention oriented covertly (no eye movement) and saccadic orienting responses both critically contribute to visual perception and involve overlapping neural substrates. In addition, recent studies show that SA modulates the orienting of spatial attention while the reverse effect, that is, the effect of spatial attention on SA, remains unsettled. In the second part of this thesis (Study 3), we aim to assess in depth the possibility of a modulatory effect of spatial attention on SA. We used a combination of the double-step target paradigm (to induce adaptation) and the cross-modal attentional-orienting paradigm to investigate the effect of tactile exogenous and endogenous spatial attention on the adaptation of reactive and voluntary saccades, respectively. Our results show significant correlations between the amount and speed of saccadic adaptive changes and the amount of attention allocated toward or away from the adapted saccade target. Thus, Study 3 brings additional arguments in favor of a coupling between spatial attention and SA, possibly by means of an effect of spatial attention on the saccadic error signals at the level of the posterior parietal cortex. Overall, this work brings additional empirical insights on the control of accuracy of non-visual RS and further highlights the role of spatial attention in SA. Even though significant advances have been seen in models investigating the nature of the error signals driving SA, they currently do not consider the coupling between spatial attention and SA. Therefore, based on the available literature and the outcomes of this thesis, we suggest that future work should take into account the role of spatial attention in error processing

APA, Harvard, Vancouver, ISO und andere Zitierweisen

9

Howarth, Christopher. „Pattern adaptation and its interocular transfer in the primary visual cortex“. Thesis, Cardiff University, 2008. http://orca.cf.ac.uk/54710/.

Der volle Inhalt der Quelle

Annotation:

Adaptation to a high contrast grating temporarily reduces the contrast sensitivity of neurons in the primary visual cortex (VI). If this adaptation is induced in one eye and the contrast tested with the other a partial transfer of the after-effect is produced, known as interocular transfer (IOT). Intrinsic hyperpolarisation of a cells membrane explains most of this effect, but not the orientation selective nature of adaptation. Optical imaging of intrinsic signals in anaesthetised cats and tree shrews was used to visualise orientation selective responses in VI before and after brief and chronic adaptation. Short term adaptation was achieved with drifting gratings of 12.5 or 50% contrast and fixed orientation (0). Three 1-sec flashes of a 100% contrast grating were used as test stimuli. 8 orientation domains were created according to orientation preference, determined on the basis of pre-adaptation orientation maps. 8 oriented test stimulus responses for each domain were obtained from the absorption signal time course averaged over all pixels. Orientation tuning curves comparable to those in single-cell experiments were produced for the orientation selective pixel populations. A region specific reduction in response was seen in the tuning curves such that responses to 0 were reduced most strongly in regions responding best to 0. An additional stimulus specific reduction was observed in responses to 6, even if 0 wasn't the optimal orientation for a domain. Chronic adaptation was induced with 1 hour of drifting sinusoidal grating in tree shrews. In contrast to a similar experiment in the cat, no alteration in the functional layout of the orientation map was observed. Extracellular recording of IOT in the cat primary visual cortex was performed to elucidate its physiological substrate. Orientation tuning curves were recorded before and after left or right eye adaptation with a 25% or 50% contrast drifting grating with the cells preferred orientation and spatial frequency. Cells were a priori categorised according to the binocularity of their control responses. Surprisingly, significant levels of IOT were observed in virtually all monocular cells. Only a weak link was found between ocular dominance and IOT in the full cell population. However, a moderate link between OD and IOT was seen in simple cells. An increase in the response to orthogonal stimuli was also seen in both monocular and binocular cells after adaptation with the non-dominant eye. A subset of complex cells did not display any IOT when adapting with the non-dominant eye and testing with the dominant eye.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

10

Shell, Jethro. „Fuzzy transfer learning“. Thesis, De Montfort University, 2013. http://hdl.handle.net/2086/8842.

Der volle Inhalt der Quelle

Annotation:

The use of machine learning to predict output from data, using a model, is a well studied area. There are, however, a number of real-world applications that require a model to be produced but have little or no data available of the specific environment. These situations are prominent in Intelligent Environments (IEs). The sparsity of the data can be a result of the physical nature of the implementation, such as sensors placed into disaster recovery scenarios, or where the focus of the data acquisition is on very defined user groups, in the case of disabled individuals. Standard machine learning approaches focus on a need for training data to come from the same domain. The restrictions of the physical nature of these environments can severely reduce data acquisition making it extremely costly, or in certain situations, impossible. This impedes the ability of these approaches to model the environments. It is this problem, in the area of IEs, that this thesis is focussed. To address complex and uncertain environments, humans have learnt to use previously acquired information to reason and understand their surroundings. Knowledge from different but related domains can be used to aid the ability to learn. For example, the ability to ride a road bicycle can help when acquiring the more sophisticated skills of mountain biking. This humanistic approach to learning can be used to tackle real-world problems where a-priori labelled training data is either difficult or not possible to gain. The transferral of knowledge from a related, but differing context can allow for the reuse and repurpose of known information. In this thesis, a novel composition of methods are brought together that are broadly based on a humanist approach to learning. Two concepts, Transfer Learning (TL) and Fuzzy Logic (FL) are combined in a framework, Fuzzy Transfer Learning (FuzzyTL), to address the problem of learning tasks that have no prior direct contextual knowledge. Through the use of a FL based learning method, uncertainty that is evident in dynamic environments is represented. By combining labelled data from a contextually related source task, and little or no unlabelled data from a target task, the framework is shown to be able to accomplish predictive tasks using models learned from contextually different data. The framework incorporates an additional novel five stage online adaptation process. By adapting the underlying fuzzy structure through the use of previous labelled knowledge and new unlabelled information, an increase in predictive performance is shown. The framework outlined is applied to two differing real-world IEs to demonstrate its ability to predict in uncertain and dynamic environments. Through a series of experiments, it is shown that the framework is capable of predicting output using differing contextual data.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

11

Bouchard, Jean-Michel. „Bimanual Transfer and Retention of Visuomotor Adaptation is Driven by Explicit Processes“. Thesis, Université d'Ottawa / University of Ottawa, 2020. http://hdl.handle.net/10393/40041.

Der volle Inhalt der Quelle

Annotation:

Reaching with altered visual feedback of the hand’s position in a virtual environment leads to reach adaptation in the trained hand, and also in the untrained hand (Wang & Sainburg, 2002). In the current study, we asked if reach adaptation in the untrained (right) hand is due to transfer of explicit (i.e., EA; conscious strategy) and/or implicit adaptation (i.e., IA; unconscious) from the left (trained) hand, and if the transfer of EA and IA changes depending on how one is made aware of the visuomotor distortion. We further asked if EA and IA are retained in the trained and untrained hand for 24 hours. Participants (n=60) were evenly divided into 3 groups (Strategy, No-Strategy, and Control). All participants reached to visual targets while seeing a cursor on the screen that was rotated 40° clockwise relative to their hand motion. Participants in the Strategy group were instructed on how to counteract the visuomotor distortion. The No-Strategy group was not told of the upcoming visuomotor distortion but was later asked to reach while engaging in any strategy they had learned in order to assess EA. Participants in the Control group were also not told of the upcoming visuomotor distortion and were never instructed to engage in any strategy when reaching. EA and IA were assessed in both the trained and untrained hands immediately following rotated reach training, and 24 hours later by having participants reach without the cursor when instructed to: (1) aim so that your hand lands on the target (to assess IA) and (2) use what was learned during training so that the cursor lands on the target (to assess EA + IA; exception of Control group). Results revealed that the groups differed with respect to the extent of reach adaptation achieved when initially training with the rotated cursor, such that the Strategy group had greater EA and less IA compared to the No-Strategy group in the trained hand. Unexpectedly, the Control group also showed less IA compared to the No-Strategy group, but was similar to the Strategy group. For both the Strategy and No-Strategy groups, EA was transferred between hands and was retained over time. While the extent of IA varied between groups in the trained (left) hand immediately following reach training trials, significant transfer of IA was not found in any of the three groups. Retention of IA was observed in the trained hand but decayed over testing days. Together, these results suggest that while initial EA and IA in the trained hand is dependent on how one is made aware of the visuomotor distortion, transfer and retention of visuomotor adaptation is driven almost exclusively by EA, regardless of instructions provided.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

12

Faucher, Marion. „Le transfert horizontal de gènes chez les mycoplasmes : de l'acquisition de l'antibiorésistance à la dynamique des génomes“. Thesis, Toulouse, INPT, 2018. http://www.theses.fr/2018INPT0117/document.

Der volle Inhalt der Quelle

Annotation:

Les mycoplasmes sont des bactéries atypiques, dépourvues de paroi et souvent considérées comme des cellules minimales du fait de la taille réduite de leur génome. De nombreuses espèces sont pathogènes et ont un impact économique important dans les filières d'élevage, notamment pour les ruminants. Les mycoplasmes n'échappent pas au phénomène mondial de la résistance aux antibiotiques. Contrairement à la plupart des autres bactéries, les mycoplasmes ne contiennent pas de plasmides conjugatifs souvent incriminés dans la dissémination horizontale de gènes de résistance, la base moléculaire principalement décrite étant la mutation chromosomique des gènes cibles. De façon générale, le transfert horizontal de gènes (HGT) chez les mycoplasmes a longtemps été sous-estimé. Récemment, deux mécanismes de HGT ont été décrits chez Mycoplasma agalactiae : le transfert d'élément conjugatif et intégratif (ICE), et le transfert non-conventionnel de régions chromosomiques par conjugaison appelé MCT (Mycoplasma Chromosomal Transfer). Nos travaux se sont attachés à explorer ce dernier mécanisme et à évaluer son impact sur l'acquisition de la résistance aux antibiotiques. Une analyse de génomique comparative a été conduite à partir du séquençage de nombreux mutants spontanément résistants et de transconjugants générés par des expériences de mating et sélectionnés pour leur résistance. Nos résultats montrent que le MCT conduit de façon distributive au transfert simultané de nombreux fragments. En une seule étape de conjugaison impliquant deux souches, ce phénomène génère une population variée de génomes hautement mosaïques. Il accélère la dissémination de l'antibiorésistance, permettant l'acquisition de plusieurs mutations distantes associées à la résistance en un seul événement. De par les multiples possibilités de réassemblage génomique qu'il produit, le MCT pourrait avoir des conséquences importantes sur d'autres processus adaptatifs comme la virulence ou la spécificité d'hôte. Enfin, les modalités distributives et l’ampleur du MCT expliquent l'origine des transferts de gènes précédemment détectés in silico entre de nombreux mycoplasmes. Ce phénomène pourrait donc avoir eu des répercussions importantes sur l'évolution de ces bactéries minimales et être un facteur clé de leur persistance et virulence actuelles
Mycoplasmas are wall-less bacteria often portrayed as minimal cells because of their reduced genomes. Several species are pathogenic and have a significant economic impact on livestock production, especially for ruminants. Mycoplasmas are also concerned with the worldwide increase in antibiotic resistance. In contrast to the majority of bacteria, these simple bacteria are deprived of conjugative plasmids that are frequently implicated in the horizontal dissemination of resistance genes: in mycoplasmas antibiotic resistance mainly relies on chromosomal mutations in target genes. In Mycoplasmas, the horizontal gene transfer (HGT) has long been underestimated. Recently, two conjugative mechanisms of HGT were described in Mycoplasma agalactiae: the transfer of an integrative and conjugative element (ICE), and the unconventional transfer of chromosomal DNA further designed by “MCT” for Mycoplasma Chromosomal Transfer. Our current study focused on exploring MCT mechanisms and on estimating its impact on antibiotic resistance dissemination. Comparative genomic analyses were performed from the sequencing (i) of spontaneous resistant mutants and (ii) of transconjugants selected by mating experiments and selected based on their resistance. Data revealed that MCT generated the simultaneous transfer of multiple, unrelated donor-fragments following a distributive process. In one conjugative step involving two strains, MCT generated a variety of highly mosaic genomes. This phenomenon was also shown to accelerate the dissemination of antibiotic resistance, by allowing in one step the acquisition of multiple and dispersed mutations associated with resistance. Due to the limitless ability of this phenomenon in reshuffling genomes, MCT may offer a valuable contribution in other adaptive processes such as virulence or host specificity. Finally, the distributive nature and the extent of MCT explain the origin of genes transfers detected in silico in several mycoplasma species. MCT is certainly a major player in the evolution of these minimal bacteria and a key factor of their persistence and virulence

APA, Harvard, Vancouver, ISO und andere Zitierweisen

13

Awang, Dzulkifli. „Potential contribution of industrial design to the adaptation of agricultural machinery for use in Malaysia“. Thesis, Manchester Metropolitan University, 1995. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.240741.

Der volle Inhalt der Quelle

Annotation:

This thesis describes a study of the potential contribution of industrial design to the adaptation of agricultural machinery for use in Malaysia. The study identifies the nature of agricultural machinery and equipment design which uses mainly adaptation concepts as a means of getting a 'quick' design. The nature of industrial design with regard to agricultural machinery and equipment design in Malaysia and the UK was also studied. This research has established the potential of industrial designers to contribute in this area of design activity. This researchu tilised a combinedm ethodologyo f'triangulation' where quantitative surveysa nd qualitative studies were used. Literature surveys, a groundwork study, questionnaire surveys, case studies, interviews, and observations were carried out. Case studies and interviews were the main methods of data collection while questionnaire surveys helped to provide information, support, and validate findings from the former methods. Case studies and interviews involved five establishmentsin Malaysia and four companiesi n the UK. Adaptation designw as confirmed as'a successfulc oncept, however, the processesin volved in the designo f machinesa nd equipmentn eedt o be developeda nd improved. There were generally no specific design processes or procedures adhered to while carrying adaptation design work. They were determinedb y the samef actors i. e. the types of projects, aims/objectiveso f project, duration and date-line, funding and level of adaptation design. From the quantitative survey and qualitatives tudyr esults,i t was establishedth at industrial design should be incorporated into the adaptation design concept. The role of industrial designers in adaptation design can be that of assisting and advising the engineers and working collectively as a team. The potential contributionso f industriald esignersw ould be to improve 'appearance('a esthetics),q uality, and overall design. Industrial design knowledge in human-factors- ergonomics/anthropometricswere also sort for in order to satisfy factors such as safety and ease of use. Industrial designers were also needed to achieve designs which could reduce costs and to increase sales of machinery/equipmenpt roducts. The thesis concludes that the diverse nature of agricultural machinery and equipment design, especiallyfo cusingo n the adaptationd esignc oncept,r equiresa synthesisa pproachr equiring the participation and consideration of such factors as: inter-disciplinary relationships, agricultural policy, designp olicy, designe ducation,a nd public awarenessA. ll thesef actors coupled with the contributions made by industrial designers would bring improvements to the success rates of adaptation design. V

APA, Harvard, Vancouver, ISO und andere Zitierweisen

14

Berglund, Eva Caroline. „Genome Evolution and Host Adaptation in Bartonella“. Doctoral thesis, Uppsala universitet, Molekylär evolution, 2009. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-108376.

Der volle Inhalt der Quelle

Annotation:

Bacteria of the genus Bartonella infect the red blood cells of a wide range of wild and domestic mammals and are transmitted between hosts by blood-sucking insects. Although most Bartonella infections are asymptomatic, the genus contains several human pathogens. In this work, host adaptation and host switches in Bartonella have been studied from a genomic perspective, with special focus on the acquisition and evolution of genes involved in host interactions. As part of this study, the complete genome of B. grahamii isolated from a Swedish wood mouse was sequenced. A genus-wide comparison revealed that rodent-associated Bartonella species, which have rarely been associated with human disease, have the largest genomes and the largest number of host-adaptability genes. Analysis of known and putative genes for host interactions identified several families of autotransporters as horizontally transferred to the Bartonella ancestor, with a possible role both during early host adaptation and subsequent host shifts. In B. grahamii, the association of a gene transfer agent (GTA) and phage-derived run-off replication of a large genomic segment was demonstrated for the first time. Among all acquisitions to the Bartonella ancestor, the only well conserved gene clusters are those that encode the GTA and contain the origin of the run-off replication. This conservation, along with a high density of host-adaptability genes in the amplified region suggest that the GTA provides a strong selective advantage, possibly by increasing recombination frequencies of host-adaptability genes, thereby facilitating evasion of the host immune system and colonization of new hosts. B. grahamii displays stronger geographic pattern and higher recombination frequencies than the cat-associated B. henselae, probably caused by different lifestyles and/or population sizes of the hosts. The genomic diversity of B. grahamii is markedly lower in Europe and North America than in Asia, possibly an effect of reduced host variability in these areas following the latest ice age.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

15

Shermin, Tasfia. „Enhancing deep transfer learning for image classification“. Thesis, Federation University Australia, 2021. http://researchonline.federation.edu.au/vital/access/HandleResolver/1959.17/179551.

Der volle Inhalt der Quelle

Annotation:

Though deep learning models require a large amount of labelled training data for yielding high performance, they are applied to accomplish many computer vision tasks such as image classification. Current models also do not perform well across different domain settings such as illumination, camera angle and real-to-synthetic. Thus the models are more likely to misclassify unknown classes as known classes. These issues challenge the supervised learning paradigm of the models and encourage the study of transfer learning approaches. Transfer learning allows us to utilise the knowledge acquired from related domains to improve performance on a target domain. Existing transfer learning approaches lack proper high-level source domain feature analyses and are prone to negative transfers for not exploring proper discriminative information across domains. Current approaches also lack at discovering necessary visual-semantic linkage and has a bias towards the source domain. In this thesis, to address these issues and improve image classification performance, we make several contributions to three different deep transfer learning scenarios, i.e., the target domain has i) labelled data; no labelled data; and no visual data. Firstly, for improving inductive transfer learning for the first scenario, we analyse the importance of high-level deep features and propose utilising them in sequential transfer learning approaches and investigating the suitable conditions for optimal performance. Secondly, to improve image classification across different domains in an open set setting by reducing negative transfers (second scenario), we propose two novel architectures. The first model has an adaptive weighting module based on underlying domain distinctive information, and the second model has an information-theoretic weighting module to reduce negative transfers. Thirdly, to learn visual classifiers when no visual data is available (third scenario) and reduce source domain bias, we propose two novel models. One model has a new two-step dense attention mechanism to discover semantic attribute-guided local visual features and mutual learning loss. The other model utilises bidirectional mapping and adversarial supervision to learn the joint distribution of source-target domains simultaneously. We propose a new pointwise mutual information dependant loss in the first model and a distance-based loss in the second one for handling source domain bias. We perform extensive evaluations on benchmark datasets and demonstrate the proposed models outperform contemporary works.
Doctor of Philosophy

APA, Harvard, Vancouver, ISO und andere Zitierweisen

16

Schmitz, Gerd [Verfasser]. „Mechanismen von visuomotorischer und audiomotorischer Adaptation: Transfer zwischen Sinnesmodalitäten und Effektoren / Gerd Schmitz“. Köln : Zentralbibliothek der Deutschen Sporthochschule, 2010. http://d-nb.info/107137933X/34.

Der volle Inhalt der Quelle

APA, Harvard, Vancouver, ISO und andere Zitierweisen

17

Butcher, Obie. „An in vivo model to assess the role of brain derived extracts on behavioural development and adaptation“. Thesis, Oxford Brookes University, 2002. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.247800.

Der volle Inhalt der Quelle

APA, Harvard, Vancouver, ISO und andere Zitierweisen

18

Al, Chalati Abdul Aziz, und Syed Asad Naveed. „Transfer Learning for Machine Diagnostics“. Thesis, Högskolan i Halmstad, Akademin för informationsteknologi, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:hh:diva-43185.

Der volle Inhalt der Quelle

Annotation:

Fault detection and diagnostics are crucial tasks in condition-based maintenance. Industries nowadays are in need of fault identification in their machines as early as possible to save money and take precautionary measures in case of fault occurrence. Also, it is beneficial for the smooth interference in the manufacturing process in which it avoids sudden malfunctioning. Having sufficient training data for industrial machines is also a major challenge which is a prerequisite for deep neural networks to train an accurate prediction model. Transfer learning in such cases is beneficial as it can be helpful in adapting different operating conditions and characteristics which is the casein real-life applications. Our work is focused on a pneumatic system which utilizes compressed air to perform operations and is used in different types of machines in the industrial field. Our novel contribution is to build upon a Domain Adversarial Neural Network (DANN) with a unique approach by incorporating ensembling techniques for diagnostics of air leakage problem in the pneumatic system under transfer learning settings. Our approach of using ensemble methods for feature extraction shows up to 5 % improvement in the performance. We have also performed a comparative analysis of our work with conventional machine and deep learning methods which depicts the importance of transfer learning and we have also demonstrated the generalization ability of our model. Lastly, we also mentioned a problem specific contribution by suggesting a feature engineering approach, such that it could be implemented on almost every pneumatic system and could potentially impact the prediction result positively. We demonstrate that our designed model with domain adaptation ability will be quite useful and beneficial for the industry by saving their time and money and providing promising results for this air leakage problem in the pneumatic system.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

19

Sutter, Christopher J. „Knowledge Transfer in Base of the Pyramid Markets“. The Ohio State University, 2013. http://rave.ohiolink.edu/etdc/view?acc_num=osu1366200889.

Der volle Inhalt der Quelle

APA, Harvard, Vancouver, ISO und andere Zitierweisen

20

Manamasa, Krishna Himaja. „Domain adaptation from 3D synthetic images to real images“. Thesis, Blekinge Tekniska Högskola, Institutionen för datavetenskap, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:bth-19303.

Der volle Inhalt der Quelle

Annotation:

Background. Domain adaptation is described as, a model learning from a source data distribution and performing well on the target data. This concept, Domain adaptation is applied to assembly-line production tasks to perform an automatic quality inspection. Objectives. The aim of this master thesis is to apply this concept of 3D domain adaptation from synthetic images to real images. It is an attempt to bridge the gap between different domains (synthetic and real point cloud images), by implementing deep learning models that learn from synthetic 3D point cloud (CAD model images) and perform well on the actual 3D point cloud (3D Camera images). Methods. Through this course of thesis project, various methods for understand- ing the data and analyzing it for bridging the gap between CAD and CAM to make them similar is looked into. Literature review and controlled experiment are research methodologies followed during implementation. In this project, we experiment with four different deep learning models with data generated and compare their performance to know which deep learning model performs best for the data. Results. The results are explained through metrics i.e, accuracy and train time, which were the outcomes of each of the deep learning models after the experiment. These metrics are illustrated in the form of graphs for comparative analysis between the models on which the data is trained and tested on. PointDAN showed better results with higher accuracy compared to the other 3 models. Conclusions. The results attained show that domain adaptation for synthetic images to real images is possible with the data generated. PointDAN deep learning model which focuses on local feature alignment and global feature alignment with single-view point data shows better results with our data.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

21

Langlois, Xavier. „Adaptation a priori et a posteriori de maillage autour d'une interface dans des problèmes thermiques“. Nancy 1, 1993. http://www.theses.fr/1993NAN10178.

Der volle Inhalt der Quelle

Annotation:

L'objet de ce travail est l'analyse mathématique et numérique d'un critère de maillage au voisinage de l'interface entre deux milieux pour un problème de transfert thermique. Nous caractérisons son domaine de validité pour diverses méthodes d'approximation numérique et nous établissons le lien avec les techniques usuelles de maillage adaptatif. Une première analyse montre que le critère est efficace sur certains schémas monodimensionnels aux différences finies à trois points en espace et qu'il se traduit par une amélioration de l'erreur de consistance au point interface. Dans le cadre de la méthode aux éléments finis de degré un, nous montrons que le critère coïncide avec celui classique d'equirepartition de la norme l#2 de l'erreur d'interpolation. Par contre, pour d'autres normes, en particulier la norme énergie, l'analyse conduit à des critères différents. Ensuite, nous situons le critère par rapport aux estimations a posteriori faites en maillage adaptatif pour des estimateurs équivalents à la norme énergie de l'erreur. Des simulations numériques illustrent tous ces aspects en dimension un et deux

APA, Harvard, Vancouver, ISO und andere Zitierweisen

22

Mottais, Angélique. „Thérapie génique non-virale de la mucoviscidose : évaluation des voies d'administration et adaptation des formulations“. Thesis, Brest, 2017. http://www.theses.fr/2017BRES0146/document.

Der volle Inhalt der Quelle

Annotation:

La mucoviscidose est une maladie génétique grave et évolutive dont les atteintes pulmonaires sont, aujourd’hui, les causes principales de décès des patients. Du fait de l’absence ou du dysfonctionnement du canal chlore CFTR, les patients atteints de mucoviscidose présentent un mucus hypervisqueux notamment au niveau des voies respiratoires. Ce mucus est un environnement favorable à la colonisation par des bactéries opportunistes telles que le Staphylococcus aureus ou le Pseudomonas aeruginosa. La chronicité des infections couplées avec une inflammation importante conduisent à la dégradation progressive des fonctions respiratoires. Actuellement, hormis la greffe coeur/poumon, aucun traitement curatif n’est encore accessible pour l’ensemble des patients. L’approche par thérapie génique apparait être une bonne stratégie pour tenter de guérir tous les patients indépendamment du type de mutations dont ils sont porteurs. Il s’agit d’apporter une copie saine du gène CFTR à l’intérieur des cellules afin que celles-ci expriment une protéine fonctionnelle. Pour ce faire, de nombreuses barrières sont à franchir. Parmi elles, la présence de bactéries dans l’environnement cellulaire est un frein s’opposant au transfert de gènes notamment par vecteurs. Il semble pertinent de développer une formulation multifonctionnelle permettant d’une part d’éliminer les bactéries en surface et d’autre part, de transfecter les cellules cibles. Cette formulation doit rester efficace après qu’elle ait été aérosolisée. Au cours de ce travail, plusieurs formulations, incorporant des lipides cationiques et des composés argent, ont été mises au point
Cystic fibrosis is a genetic disease with lung damages as the current main causes of death. Due to the absence or dysfunction of the CFTR chloride channel, CF patients have hyper-viscous mucus, particularly in the respiratory tract. This mucus is an environment favorable to infection development by opportunistic bacteria such as Staphylococcus aureus or Pseudomonas aeruginosa. The chronicity of infections coupled with significant inflammation leads to the progressive degradation of respiratory functions. Currently, apart from the heart-lung transplantation, no cure is still available for all patients.The approach by gene therapy appears to be a good strategy to cure all patients regardless of the type of mutations they have. It is a matter of bringing a healthy copy of the CFTR gene into the cells so that they express a functional protein. To do this, many barriers must be overcome. Among them, the presence of bacteria in the cellular environment is a brake against the transfer of genes in particular by vectors. It seems pertinent to develop a multifunctional formulation that on the one hand eliminates surface bacteria and on the other hand transfect the target cells. This formulation must remain effective after it has been aerosolized. During this work, several formulations, incorporating cationic lipids and silver compounds, have been developed

APA, Harvard, Vancouver, ISO und andere Zitierweisen

23

Voo, Charles. „Management of low and variable bit rate ATM Adaptation Layer Type 2 traffic“. University of Western Australia. School of Electrical, Electronic and Computer Engineering, 2003. http://theses.library.uwa.edu.au/adt-WU2004.0030.

Der volle Inhalt der Quelle

Annotation:

Asynchronous Transfer Mode (ATM) Adaptation Layer Type 2 (AAL2) has been developed to carry low and variable bit rate traffic. It provides high bandwidth efficiency with low packing delay by allowing voice traffic from different AAL2 channels to be multiplexed onto a single ATM virtual channel connection. Examples of where AAL2 are used include the Code Division Multiple Access and the Third Generation mobile telephony networks. The main objective of this thesis is to study traditional and novel AAL2 multiplexing methods and to characterise their performance when carrying low and variable bit rate (VBR) voice traffic. This work develops a comprehensive QoS framework which is used as a basis to study the performance of the AAL2 multiplexer system. In this QoS framework the effects of packet delay, delay variation, subjective voice quality and bandwidth utilisation are all used to determine the overall performance of the end-to-end system for the support of real time voice communications. Extensions to existing AAL2 voice multiplexers are proposed and characterised. In the case where different types of voice applications are presented to the AAL2 multiplexer, it was observed that increased efficiency gains are possible when a priority queuing scheme is introduced into the traditional AAL2 multiplexer system. Studies of the voice traffic characteristics and their effects on the performance of the AAL2 multiplexer are also investigated. It is shown that particular source behaviours can have deleterious effect on the performance of the AAL2 multiplexer. Methods of isolating these voice sources are examined and the performance of the AAL2 multiplexer re-evaluated to show the beneficial effects of a particular source isolation technique. The extent to which statistical multiplexing is possible for real time variable VBR sources is theoretically examined. These calculations highlight the difficulties in multiplexing VBR real time traffic while maintaining guaranteed delay bounds for these sources. Based on these calculations, multiplexing schemes that incorporate data transfers within the real time traffic transfer are proposed as alternatives for utilising unused bandwidth caused by the VBR nature of the voice traffic.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

24

Fernandes, Montesuma Eduardo. „Multi-Source Domain Adaptation through Wasserstein Barycenters“. Electronic Thesis or Diss., université Paris-Saclay, 2024. http://www.theses.fr/2024UPASG045.

Der volle Inhalt der Quelle

Annotation:

Les systèmes d'apprentissage automatique fonctionnent sous l'hypothèse que les conditions d'entraînement et de test ne changent pas. Néanmoins, cette hypothèse est rarement vérifiée en pratique. En conséquence, le système est entraîné avec des données qui ne sont plus représentatives des données sur lesquelles il sera testé : la mesure de probabilité des données évolue entre les périodes d'entraînement et de test. Ce scénario est connu dans la littérature sous le nom de décalage de distribution entre deux domaines : une source et une cible. Une généralisation évidente de ce problème considère que les données d'entraînement présentent elles-mêmes plusieurs décalages intrinsèques. On parle, donc, d'adaptation de domaine à sources multiples (MSDA). Dans ce contexte, le transport optimal est un outil de mathématique utile. En particulier, qui sert pour comparer et manipuler des mesures de probabilité. Cette thèse étudie les contributions du transport optimal à l'adaptation de domaines à sources multiples. Nous le faisons à travers des barycentres de Wasserstein, un objet qui définit une moyenne pondérée, dans l'espace des mesures de probabilité, des multiples domaines en MSDA. Basé sur ce concept, nous proposons : (i) une nouvelle notion de barycentre lorsque les mesures en question sont étiquetées, (ii) un nouveau problème d'apprentissage de dictionnaire sur des mesures de probabilité empiriques et (iii) de nouveaux outils pour l'adaptation de domaines via le transport optimal de modèles de mélanges Gaussiens. Nos méthodes améliorent les performances de l'adaptation de domaines par rapport aux méthodes existantes utilisant le transport optimal sur des benchmarks d'images et de diagnostic de défauts inter-domaines. Notre travail ouvre une perspective de recherche intéressante sur l'apprentissage de l'enveloppe barycentrique de mesures de probabilité
Machine learning systems work under the assumption that training and test conditions are uniform, i.e., they do not change. However, this hypothesis is seldom met in practice. Hence, the system is trained with data that is no longer representative of the data it will be tested on. This case is represented by a shift in the probability measure generating the data. This scenario is known in the literature as distributional shift between two domains: a source, and a target. A straightforward generalization of this problem is when training data itself exhibit shifts on its own. In this case, one consider Multi Source Domain Adaptation (MSDA). In this context, optimal transport is an useful field of mathematics. Especially, optimal transport serves as a toolbox, for comparing and manipulating probability measures. This thesis studies the contributions of optimal transport to multi-source domain adaptation. We do so through Wasserstein barycenters, an object that defines a weighted average, in the space of probability measures, for the multiple domains in MSDA. Based on this concept, we propose: (i) a novel notion of barycenter, when the measures at hand are equipped with labels, (ii) a novel dictionary learning problem over empirical probability measures and (iii) new tools for domain adaptation through the optimal transport of Gaussian mixture models. Through our methods, we are able to improve domain adaptation performance in comparison with previous optimal transport-based methods on image, and cross-domain fault diagnosis benchmarks. Our work opens an interesting research direction, on learning the barycentric hull of probability measures

APA, Harvard, Vancouver, ISO und andere Zitierweisen

25

Bose, Tulika. „Transfer learning for abusive language detection“. Electronic Thesis or Diss., Université de Lorraine, 2023. http://www.theses.fr/2023LORR0019.

Der volle Inhalt der Quelle

Annotation:

La prolifération des médias sociaux, malgré ses nombreux avantages, a entraîné une augmentation des propos injurieux. Ces propos, qui sont généralement blessants, toxiques ou empreints de préjugés à l'encontre d'individus ou de groupes, doivent être détectés et modérés rapidement par les plateformes en ligne. Les modèles d'apprentissage profond pour la détection de propos abusifs ont montré des niveaux de performance élevé quand ils sont évalués sur des données similaires à celles qui ont servi à entraîner les modèles, mais sont nettement moins performants s'ils sont évalués sur des données dont la distribution est différente. En outre, ils nécessitent une quantité considérable de données étiquetées coûteuses pour l'apprentissage. C'est pour cela qu'il est intéressant d'étudier le transfert efficace de connaissances à partir de corpus annotés existants de propos abusifs. Cette thèse étudie le problème de l'apprentissage par transfert pour la détection de propos abusifs et explore diverses solutions pour améliorer le transfert de connaissances dans des scénarios inter corpus.Tout d'abord, nous analysons la généralisabilité inter-corpus des modules de détection de propos abusifs sans accéder à des données cibles pendant le processus d'apprentissage. Nous examinons si la combinaison des représentations issues du thème (topic) avec des représentations contextuelles peut améliorer la généralisabilité. Nous montrons que l'association de commentaires du corpus cible avec des thèmes du corpus d'entraînement peut fournir des informations complémentaires pour un meilleur transfert inter-corpus.Ensuite, nous explorons l'adaptation au domaine non supervisée (UDA, Unsupervised Domain Adaptation), un type d'apprentissage par transfert transductif, avec accès au corpus cible non étiqueté. Nous explorons certaines approches UDA populaires dans la classification des sentiments pour la détection de propos abusifs dans le cadre de corpus croisés. Nous adaptons ensuite une variante du modèle BERT au corpus cible non étiqueté en utilisant la technique du modèle de langue avec masques (MLM Masked Language Model). Alors que cette dernière améliore les performances inter-corpus, les autres approches UDA ont des performances sous-optimales. Notre analyse révèle leurs limites et souligne le besoin de méthodes d'adaptation efficaces pour cette tâche.Comme troisième contribution, nous proposons deux approches d'adaptation au domaine utilisant les attributions de caractéristiques (feature attributions), qui sont des explications a posteriori du modèle. En particulier, nous étudions le problème des corrélations erronées (spurious correlations) spécifiques à un corpus qui limitent la généralisation pour la détection des discours de haine, un sous-ensemble des propos abusifs. Alors que les approches de la littérature reposent sur une liste de termes établie manuellement, nous extrayons et pénalisons automatiquement les termes qui causent des corrélations erronées. Nos approches dynamiques améliorent les performances dans le cas de corpus croisés par rapport aux travaux précédents, à la fois indépendamment et en combinaison avec des dictionnaires prédéfinis.Enfin, nous considérons le transfert de connaissances d'un domaine source avec beaucoup de données étiquetées vers un domaine cible, où peu d'instances étiquetées sont disponibles. Nous proposons une nouvelle stratégie d'apprentissage, qui permet une modélisation flexible de la proximité relative des voisins récupérés dans le corpus source pour apprendre la quantité de transfert utile. Nous incorporons les informations de voisinage avec une méthode de transport optimal (Optimal Transport ) qui exploite la géométrie de l'espace de représentation (embedding space) . En alignant les distributions conjointes de l'embedding et des étiquettes du voisinage, nous montrons des améliorations substantielles dans des corpus de discours haineux de taille réduite
The proliferation of social media, despite its multitude of benefits, has led to the increased spread of abusive language. Such language, being typically hurtful, toxic, or prejudiced against individuals or groups, requires timely detection and moderation by online platforms. Deep learning models for detecting abusive language have displayed great levels of in-corpus performance but underperform substantially outside the training distribution. Moreover, they require a considerable amount of expensive labeled data for training.This strongly encourages the effective transfer of knowledge from the existing annotated abusive language resources that may have different distributions to low-resource corpora. This thesis studies the problem of transfer learning for abusive language detection and explores various solutions to improve knowledge transfer in cross-corpus scenarios.First, we analyze the cross-corpus generalizability of abusive language detection models without accessing the target during training. We investigate if combining topic model representations with contextual representations can improve generalizability. The association of unseen target comments with abusive language topics in the training corpus is shown to provide complementary information for a better cross-corpus transfer.Secondly, we explore Unsupervised Domain Adaptation (UDA), a type of transductive transfer learning, with access to the unlabeled target corpus. Some popular UDA approaches from sentiment classification are analyzed for cross-corpus abusive language detection. We further adapt a BERT model variant to the unlabeled target using the Masked Language Model (MLM) objective. While the latter improves the cross-corpus performance, the other UDA methods perform sub-optimally. Our analysis reveals their limitations and emphasizes the need for effective adaptation methods suited to this task.As our third contribution, we propose two DA approaches using feature attributions, which are post-hoc model explanations. Particularly, the problem of spurious corpus-specific correlations is studied that restrict the generalizability of classifiers for detecting hate speech, a sub-category of abusive language. While the previous approaches rely on a manually curated list of terms, we automatically extract and penalize the terms causing spurious correlations. Our dynamic approaches improve the cross-corpus performanceover previous works both independently and in combination with pre-defined dictionaries.Finally, we consider transferring knowledge from a resource-rich source to a low-resource target with fewer labeled instances, across different online platforms. A novel training strategy is proposed, which allows flexible modeling of the relative proximity of neighbors retrieved from the resource-rich corpus to learn the amount of transfer. We incorporate neighborhood information with Optimal Transport that permits exploitingthe embedding space geometry. By aligning the joint embedding and label distributions of neighbors, substantial improvements are obtained in low-resource hate speech corpora

APA, Harvard, Vancouver, ISO und andere Zitierweisen

26

Choi, Jin-Woo. „Action Recognition with Knowledge Transfer“. Diss., Virginia Tech, 2021. http://hdl.handle.net/10919/101780.

Der volle Inhalt der Quelle

Annotation:

Recent progress on deep neural networks has shown remarkable action recognition performance from videos. The remarkable performance is often achieved by transfer learning: training a model on a large-scale labeled dataset (source) and then fine-tuning the model on the small-scale labeled datasets (targets). However, existing action recognition models do not always generalize well on new tasks or datasets because of the following two reasons. i) Current action recognition datasets have a spurious correlation between action types and background scene types. The models trained on these datasets are biased towards the scene instead of focusing on the actual action. This scene bias leads to poor generalization performance. ii) Directly testing the model trained on the source data on the target data leads to poor performance as the source, and target distributions are different. Fine-tuning the model on the target data can mitigate this issue. However, manual labeling small- scale target videos is labor-intensive. In this dissertation, I propose solutions to these two problems. For the first problem, I propose to learn scene-invariant action representations to mitigate the scene bias in action recognition models. Specifically, I augment the standard cross-entropy loss for action classification with 1) an adversarial loss for the scene types and 2) a human mask confusion loss for videos where the human actors are invisible. These two losses encourage learning representations unsuitable for predicting 1) the correct scene types and 2) the correct action types when there is no evidence. I validate the efficacy of the proposed method by transfer learning experiments. I trans- fer the pre-trained model to three different tasks, including action classification, temporal action localization, and spatio-temporal action detection. The results show consistent improvement over the baselines for every task and dataset. I formulate human action recognition as an unsupervised domain adaptation (UDA) problem to handle the second problem. In the UDA setting, we have many labeled videos as source data and unlabeled videos as target data. We can use already exist- ing labeled video datasets as source data in this setting. The task is to align the source and target feature distributions so that the learned model can generalize well on the target data. I propose 1) aligning the more important temporal part of each video and 2) encouraging the model to focus on action, not the background scene, to learn domain-invariant action representations. The proposed method is simple and intuitive while achieving state-of-the-art performance without training on a lot of labeled target videos. I relax the unsupervised target data setting to a sparsely labeled target data setting. Then I explore the semi-supervised video action recognition, where we have a lot of labeled videos as source data and sparsely labeled videos as target data. The semi-supervised setting is practical as sometimes we can afford a little bit of cost for labeling target data. I propose multiple video data augmentation methods to inject photometric, geometric, temporal, and scene invariances to the action recognition model in this setting. The resulting method shows favorable performance on the public benchmarks.
Doctor of Philosophy
Recent progress on deep learning has shown remarkable action recognition performance. The remarkable performance is often achieved by transferring the knowledge learned from existing large-scale data to the small-scale data specific to applications. However, existing action recog- nition models do not always work well on new tasks and datasets because of the following two problems. i) Current action recognition datasets have a spurious correlation between action types and background scene types. The models trained on these datasets are biased towards the scene instead of focusing on the actual action. This scene bias leads to poor performance on the new datasets and tasks. ii) Directly testing the model trained on the source data on the target data leads to poor performance as the source, and target distributions are different. Fine-tuning the model on the target data can mitigate this issue. However, manual labeling small-scale target videos is labor-intensive. In this dissertation, I propose solutions to these two problems. To tackle the first problem, I propose to learn scene-invariant action representations to mitigate background scene- biased human action recognition models for the first problem. Specifically, the proposed method learns representations that cannot predict the scene types and the correct actions when there is no evidence. I validate the proposed method's effectiveness by transferring the pre-trained model to multiple action understanding tasks. The results show consistent improvement over the baselines for every task and dataset. To handle the second problem, I formulate human action recognition as an unsupervised learning problem on the target data. In this setting, we have many labeled videos as source data and unlabeled videos as target data. We can use already existing labeled video datasets as source data in this setting. The task is to align the source and target feature distributions so that the learned model can generalize well on the target data. I propose 1) aligning the more important temporal part of each video and 2) encouraging the model to focus on action, not the background scene. The proposed method is simple and intuitive while achieving state-of-the-art performance without training on a lot of labeled target videos. I relax the unsupervised target data setting to a sparsely labeled target data setting. Here, we have many labeled videos as source data and sparsely labeled videos as target data. The setting is practical as sometimes we can afford a little bit of cost for labeling target data. I propose multiple video data augmentation methods to inject color, spatial, temporal, and scene invariances to the action recognition model in this setting. The resulting method shows favorable performance on the public benchmarks.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

27

Azamfar, Moslem. „Deep Learning-based Domain Adaptation Methodology for Fault Diagnosis of Complex Manufacturing Systems“. University of Cincinnati / OhioLINK, 2021. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1623168030554759.

Der volle Inhalt der Quelle

APA, Harvard, Vancouver, ISO und andere Zitierweisen

28

Burcham, Catherine Elizabeth. „The experience of male adolescent refugees during their transfer and adaptation to a UK secondary school“. Thesis, University College London (University of London), 2009. http://discovery.ucl.ac.uk/10019926/.

Der volle Inhalt der Quelle

Annotation:

Making a successful transfer to secondary school in the UK has been linked to a range of positive outcomes for refugee children. Yet research investigating this experience from the perspective of refugee children themselves is scarce. This study aimed to gain an in-depth understanding of how male adolescent refugees experienced their transfer and adaptation to a secondary school in the UK. The research used a qualitative design, it was idiographic arid the approach adopted was Interpretative Phenomenological Analysis (IPA). Semi-structured interviews were completed with six male adolescent refugees. The data generated three superordinate themes which reflected the participants' sense of being in need of help during the early stages of their transfer, their process of adapting to school and developing a sense of belonging in this context, and their overriding need to feel safe. Participants identified a wide range of factors as supporting and hindering their transfer and adaptation to secondary school. These are explored in relation to existing research and psychological theories. Implications for the practice of Educational Psychologists and schools are offered. In particular, results showed that older siblings were often the primary educators of participants at home. Schools that adopt a family engagement framework, rather than parental engagement, may more successfully support the learning of refugee children. All participants experienced a strong motivation to learn English and believed that speaking their first language in school would hinder this process. Further research is needed to investigate perceptions of first language use in school. Finally there is a reflection upon the applicability of using IPA with children learning English as an additional language. The author suggests that IPA can be successfully applied in research with this population.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

29

Wulfmeier, Markus. „Efficient supervision for robot learning via imitation, simulation, and adaptation“. Thesis, University of Oxford, 2018. http://ora.ox.ac.uk/objects/uuid:2b5eeb55-639a-40ae-83b7-bd01fc8fd6cc.

Der volle Inhalt der Quelle

Annotation:

In order to enable more widespread application of robots, we are required to reduce the human effort for the introduction of existing robotic platforms to new environments and tasks. In this thesis, we identify three complementary strategies to address this challenge, via the use of imitation learning, domain adaptation, and transfer learning based on simulations. The overall work strives to reduce the effort of generating training data by employing inexpensively obtainable labels and by transferring information between different domains with deviating underlying properties. Imitation learning enables a straightforward way for untrained personnel to teach robots to perform tasks by providing demonstrations, which represent a comparably inexpensive source of supervision. We develop a scalable approach to identify the preferences underlying demonstration data via the framework of inverse reinforcement learning. The method enables integration of the extracted preferences as cost maps into existing motion planning systems. We further incorporate prior domain knowledge and demonstrate that the approach outperforms the baselines including manually crafted cost functions. In addition to employing low-cost labels from demonstration, we investigate the adaptation of models to domains without available supervisory information. Specifically, the challenge of appearance changes in outdoor robotics such as illumination and weather shifts is addressed using an adversarial domain adaptation approach. A principal advantage of the method over prior work is the straightforwardness of adapting arbitrary, state-of-the-art neural network architectures. Finally, we demonstrate performance benefits of the method for semantic segmentation of drivable terrain. Our last contribution focuses on simulation to real world transfer learning, where the characteristic differences are not only regarding the visual appearance but the underlying system dynamics. Our work aims at parallel training in both systems and mutual guidance via auxiliary alignment rewards to accelerate training for real world systems. The approach is shown to outperform various baselines as well as a unilateral alignment variant.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

30

King, Judy A. „Adaptation of Striped Bass to Sea Water Following Direct Transfer from Freshwater: Morphological, Biochemical, and Physiological Parameters“. Digital Commons @ East Tennessee State University, 1987. https://dc.etsu.edu/etd/2932.

Der volle Inhalt der Quelle

Annotation:

There has been heightened interest in the biology of striped bass (Morone saxatilis) because of increased pollution in their native spawning grounds and because of their extensive use in landlocked sport fisheries. Their euryhalinity makes them an excellent species for osmoregulation studies. The objective of this research was to study the rate of adaptation of striped bass gills to sea water (3% salt) after direct transfer from freshwater using biochemical (ion transport enzyme levels), physiological (chloride efflux), and ultrastructural methods. Striped bass have specialized osmoregulatory cells located on the interlamellar and afferent surfaces of their gill filaments as shown by light microscopy (LM), transmission electron microscopy (TEM), and scanning electron microscopy (SEM). SEM studies show that apical pit (opening of one or more chloride cells) morphology changes during sea water adaptation, and the number of apical pits increases by 32.5% after two weeks in sea water. Chloride cell size and number, extent of basolateral tubular system, and number of mitochondria per chloride cell appear to increase upon adaptation to sea water. Sodium-potassium adenosine triphosphatase (Na,K-ATPase) activity is maximal on day 3 after transfer to sea water. Studies suggest that cortisol may act as a hormonal mediator for long term adaptation to sea water. The general morphology of both freshwater and sea water adapted fish gills were studied. Preliminary studies indicate that the osmium-dimethylsulfoxide-osmium method can be used to investigate intracellular structural changes in striped bass gills. Since the chloride cells are associated with the afferent surface of the filament, the blood supply to that area is also of great interest in osmoregulation studies. Studies of the gill vasculature using corrosion casting (i.e. filling blood vessels with plastic resins) and SEM indicate that the blood vessel distribution in the striped bass gill is similar to that of other euryhaline species with arterio-arterial, arterio-venous, and nutritive pathways. Blood flow may be controlled at a variety of places by sphincters, shunts and cellular contraction. Correlation of these biochemical, physiological and anatomical measurements will aid in the understanding of the process of adaptation to sea water. (Abstract shortened with permission of author.)

APA, Harvard, Vancouver, ISO und andere Zitierweisen

31

Caro, Quintero Alejandro. „The role of horizontal gene transfer in bacterial evolution“. Diss., Georgia Institute of Technology, 2013. http://hdl.handle.net/1853/48979.

Der volle Inhalt der Quelle

Annotation:

Horizontal gene transfer (HGT) is probably the most important mechanism for functional novelty and adaption in bacteria. However, a robust understanding of the rates of HGT for most bacterial species and the influence of the ecological settings on the rates remain elusive. Four whole-genome comparative studies of free-living bacteria will be described that integrated physiological and ecological data with novel detection bioinformatic pipelines to elucidate the magnitude of HGT at three distinct levels of genetic relatedness: i) the species level, where overlapping ecological niche among co-occurring bacteria in the water column of the Baltic Sea has caused HGT to have been so rampant that it has served as the force of species cohesion; ii) the genus level, where HGT appeared to predominantly mobilize a limited number of genes with ecological/selective advantage (e.g., antibiotic resistance genes) among distinct pathogenic Campylobacter species and hence, did not lead to species convergence; and iii) the phylum level, where HGT was found to be, in general, less frequent than the genus level but, over evolutionary time, has assembled a large part of the metabolic functions of natural microbial communities, especially within organic matter rich, anaerobic habitats. In conclusion, this work advances the methods to link ecological relationships with HGT and suggests that HGT among very divergent organisms may have been more frequent than previously thought and led to successful adaptation.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

32

Koop, Julie L. „A new strain of Wolbachia in the harlequin beetle riding pseudoscorpion male killing, reproductive compensation and horizontal gene transfer /“. abstract, 2008. http://0-gateway.proquest.com.innopac.library.unr.edu/openurl?url_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation&res_dat=xri:pqdiss&rft_dat=xri:pqdiss:1460764.

Der volle Inhalt der Quelle

APA, Harvard, Vancouver, ISO und andere Zitierweisen

33

Tamyez, Bajuri M. H. „An examination of the potential of industrial designers to contribute towards the process of adaptation of technology transferred from the United Kingdom to Malaysia“. Thesis, Manchester Metropolitan University, 1988. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.379690.

Der volle Inhalt der Quelle

APA, Harvard, Vancouver, ISO und andere Zitierweisen

34

Songpo, Li, und Zhang Dong. „Human Resources Management Practices in a Cross Cultural Environment : Bank of China“. Thesis, Högskolan i Gävle, Avdelningen för ekonomi, 2013. http://urn.kb.se/resolve?urn=urn:nbn:se:hig:diva-15477.

Der volle Inhalt der Quelle

Annotation:

Aim: With the tendency of globalization, the field of the human resource management (HRM) in multinational companies (MNC) becomes a heated topic. Being interested in this field, we choose Bank of China (BOC) to investigate what factors influence the process of transfer and adaptation of HRM practices for a Chinese bank in a cross cultural environment. Method: This work has been adopted qualitative method as the primary data by interviewed six respondents of managers and employees from BOC in China and Sweden by means of the face-to-face interview, e-mail, Skype and the online instant messaging software (QQ). Extant literature, books and online resources are the secondary data. Result & Conclusions: The result reveals that when transferring and adapting the HRM practices in a foreign country, MNCs are influenced by national and company level factors. At the national level, national culture and laws and regulations in the targeted country are the primary aspects while corporate culture of the targeted company and the strategic goals of the branches are the main factors at the company level. Suggestions for future research: This study only focuses on four main factors affecting the process of transfer and adaption of HRM process, which may not contain all factors. Besides, it is based on a case study with qualitative data, thus the research areas can be expanded to a broader domain and the research methods can be used in both qualitative and quantitative approaches. Last but not least, staffing performance can be a very interesting field to be investigated. Contribution of the thesis: We contribute with a model based on four factors to identify the process of transfer and adaptation of HRM practices in a MNC. We also make a unique contribution to empirical study of the process of HRM practices of a Chinese bank entering to Sweden. Finally, this model can be utilized as a tool of other MNCs when implementing HRM practices from the home country to the host country.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

35

Zijlstra, Tim. „The role of social media in relation to knowledge transfer and professional development“. Thesis, Loughborough University, 2016. https://dspace.lboro.ac.uk/2134/21265.

Der volle Inhalt der Quelle

Annotation:

Introduction: Professional development is an important aspect of working practices for academic librarians. In the past decades the academic library has been subjected to an enormous range of technological and cultural advancements that have constantly required its staff to develop their professional knowledge and understanding. One of the most recent revolutions has been the advent of social media. This advent of a new technology can provide challenges and opportunities but to fully understand these we need to analyse the relationship between social media and various context in more depth. Method: This inductive grounded theory study was iteratively conducted in three different academic libraries in England. At each location an analysis of the social media in use both organisationally and individually was made, and a purposefully selected number of actors was interviewed to gain a deeper understanding of their ideas about social media, CPD and the potential to learn from social media. Findings: The findings demonstrate that the managerial view of CPD is not necessarily compatible with the perception, or needs, of practitioners in relation to their professional development. To increase understanding of this domain, the study analysed the underlying factors related to both social media use and perception, and participation in and perception of professional development. This has led to the discovery of a theory that can be summarised as Continuous Professional Adaptation is learning that occurs as a consequence of professional awareness and preparedness through the use of informal networks. Outcome: The grounded theory presented in this document demonstrates a need to expand the dialogue on professional development in academic libraries, in both an academic and practical context, based on the advent of social media. Formal development in the shape of Continuing Professional Development (CPD) is still regarded as the most relevant method of developing human resources due to the prevalence of a managerial approach to professional development. Conclusion: Increasingly academics and practitioners require more than the formal approach to professional development which can be classed as increasingly being irrelevant due to practical limitations of delivery time and mode and speed of changes occurring. This research contributes to that dialogue, providing a unique approach by incorporating social media with professional development.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

36

Sadeghi, Mohsen Mir Mohammad. „Peer relations and school adjustment : a longitudinal study of Iranian adolescents' adaptation after transfer to a new school“. Thesis, UCL Institute of Education (IOE), 2012. http://eprints.ioe.ac.uk/20703/.

Der volle Inhalt der Quelle

Annotation:

This thesis is based on a longitudinal study of the Iranian pre-adolescents entering junior high school in Iran. It looks at the effect of peer relationships on school adjustment using a non-experimental design. The importance of adjustment to school and its consequences to a child's development has been documented in previous research. Moreover, it can be argued that school adjustment becomes even more essential after transfer to a new school. One of the major contributors to school adjustment is the child's relationship with peers. Studies on peer relationships consider three major aspects for that; friendship, sociometric status and relationships in cliques. However, such aspects have seldom been studied simultaneously. Additionally, a considerable body of literature exists concerning social networks that can be successfully utilised to examine children's relationships in the school. Social networks provide a good opportunity to develop an understanding of children's relationships. This study endeavours to determine associations between the aspects of a child's relationships and adjustment to school using some terminology and methods from the social networks literature. This study draws upon data on 389 participants from first year of junior high schools in Iran. The data was collected at the beginning of the 2008-09 school year and again at the conclusion of the school year. The longitudinal design of the study facilitated the comparison between the two timeframes and the opportunity to measure the effects of the three relationship types on different aspects of school adjustment. Results depict meaningful associations between the two domains of this study (i.e. peer relations and school adjustment) and emphasize the role that children's peer relations can have on their adaption. The study also reveals the importance of including different types of peer relations collectively in the research. Implications for practice and direction for future research are suggested based on the findings.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

37

Hall, John Joseph. „T.B. Direct Observation of Treatment, Short Course (DOTS) - Political Commitment, Policy Transfer and Adaptation in Papua New Guinea“. Thesis, The University of Sydney, 2015. http://hdl.handle.net/2123/14753.

Der volle Inhalt der Quelle

Annotation:

This thesis examines the Tuberculosis Directly Observed Treatment, Short Course (DOTS) policy in Papua New Guinea. It specifically examines the 2006 PNG TB DOTS policy process and how the complex and particular policy context in PNG is addressed, and the role of international donors, in the policy agenda setting process and policy content. A particular focus is the meaning of TB DOTS Policy Component 1 – Political Commitment, and how this shapes the governance structures and health system governance in PNG which are a central part of the policy context for the implementation of a national TB DOTS policy in a country with a continuing high burden of disease from TB.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

38

Peyrache, Jean-Philippe. „Nouvelles approches itératives avec garanties théoriques pour l'adaptation de domaine non supervisée“. Thesis, Saint-Etienne, 2014. http://www.theses.fr/2014STET4023/document.

Der volle Inhalt der Quelle

Annotation:

Ces dernières années, l’intérêt pour l’apprentissage automatique n’a cessé d’augmenter dans des domaines aussi variés que la reconnaissance d’images ou l’analyse de données médicales. Cependant, une limitation du cadre classique PAC a récemment été mise en avant. Elle a entraîné l’émergence d’un nouvel axe de recherche : l’Adaptation de Domaine, dans lequel on considère que les données d’apprentissage proviennent d’une distribution (dite source) différente de celle (dite cible) dont sont issues les données de test. Les premiers travaux théoriques effectués ont débouché sur la conclusion selon laquelle une bonne performance sur le test peut s’obtenir en minimisant à la fois l’erreur sur le domaine source et un terme de divergence entre les deux distributions. Trois grandes catégories d’approches s’en inspirent : par repondération, par reprojection et par auto-étiquetage. Dans ce travail de thèse, nous proposons deux contributions. La première est une approche de reprojection basée sur la théorie du boosting et s’appliquant aux données numériques. Celle-ci offre des garanties théoriques intéressantes et semble également en mesure d’obtenir de bonnes performances en généralisation. Notre seconde contribution consiste d’une part en la proposition d’un cadre permettant de combler le manque de résultats théoriques pour les méthodes d’auto-étiquetage en donnant des conditions nécessaires à la réussite de ce type d’algorithme. D’autre part, nous proposons dans ce cadre une nouvelle approche utilisant la théorie des (epsilon, gamma, tau)-bonnes fonctions de similarité afin de contourner les limitations imposées par la théorie des noyaux dans le contexte des données structurées
During the past few years, an increasing interest for Machine Learning has been encountered, in various domains like image recognition or medical data analysis. However, a limitation of the classical PAC framework has recently been highlighted. It led to the emergence of a new research axis: Domain Adaptation (DA), in which learning data are considered as coming from a distribution (the source one) different from the one (the target one) from which are generated test data. The first theoretical works concluded that a good performance on the target domain can be obtained by minimizing in the same time the source error and a divergence term between the two distributions. Three main categories of approaches are derived from this idea : by reweighting, by reprojection and by self-labeling. In this thesis work, we propose two contributions. The first one is a reprojection approach based on boosting theory and designed for numerical data. It offers interesting theoretical guarantees and also seems able to obtain good generalization performances. Our second contribution consists first in a framework filling the gap of the lack of theoretical results for self-labeling methods by introducing necessary conditions ensuring the good behavior of this kind of algorithm. On the other hand, we propose in this framework a new approach, using the theory of (epsilon, gamma, tau)- good similarity functions to go around the limitations due to the use of kernel theory in the specific context of structured data

APA, Harvard, Vancouver, ISO und andere Zitierweisen

39

Tao, Wang. „Adapting multiple datasets for better mammography tumor detection“. Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-231867.

Der volle Inhalt der Quelle

Annotation:

In Sweden, women of age between of 40 and 74 go through regular screening of their breasts every 18-24 months. The screening mainly involves obtaining a mammogram and having radiologists analyze them to detect any sign of breast cancer. However reading a mammography image requires experienced radiologist, and the lack of radiologist reduces the hospital's operating efficiency. What's more, mammography from different facilities increases the difficulty of diagnosis. Our work proposed a deep learning segmentation system which could adapt to mammography from various facilities and locate the position of the tumor. We train and test our method on two public mammography datasets and do several experiments to find the best parameter setting for our system. The test segmentation results suggest that our system could play as an auxiliary diagnosis tool for breast cancer diagnosis and improves diagnostic accuracy and efficiency.
I Sverige går kvinnor i åldrarna mellan 40 och 74 igenom regelbunden screening av sina bröst med 18-24 månaders mellanrum. Screeningen innbär huvudsakligen att ta mammogram och att låta radiologer analysera dem för att upptäcka tecken på bröstcancer. Emellertid krävs det en erfaren radiolog för att tyda en mammografibild, och bristen på radiologer reducerar sjukhusets operativa effektivitet. Dessutom, att mammografin kommer från olika anläggningar ökar svårigheten att diagnostisera. Vårt arbete föreslår ett djuplärande segmenteringssystem som kan anpassa sig till mammografi från olika anläggningar och lokalisera tumörens position. Vi tränar och testar vår metod på två offentliga mammografidataset och gör flera experiment för att hitta den bästa parameterinställningen för vårt system. Testsegmenteringsresultaten tyder på att vårt system kan fungera som ett hjälpdiagnosverktyg vid diagnos av bröstcancer och förbättra diagnostisk noggrannhet och effektivitet.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

40

Ruh, Mylene. „Déterminisme de la spécificité d'hôte et rôle des effecteurs TAL dans l'interaction Xanthomonas - haricot“. Thesis, Rennes, Agrocampus Ouest, 2017. http://www.theses.fr/2017NSARC131/document.

Der volle Inhalt der Quelle

Annotation:

La graisse commune est la principale phytobactériose du haricot.Cette maladie est causée par Xanthomonas citri pv. fuscans(Xcf) et X. phaseoli pv. phaseoli (Xpp). Xcf et Xpp étantphylogénétiquement distantes, leur capacité à produire lesmêmes symptômes sur haricot serait le fruit d’une convergencepathologique. L’objectif de cette thèse était d’identifi erles gènes candidats pour la spécifi cité d’hôte et d’étudier le rôledes effecteurs Transcription Activator-Like (TAL) dans l’interactionXanthomonas - haricot. Par une approche de génomiquecomparative, nous avons identifi é 116 gènes spécifi ques desagents de la graisse commune, dont un grand nombre a ététransféré horizontalement entre Xcf et Xpp. Ces gènes codentdes protéines intervenant aux différentes étapes de l’interaction.L’obtention du génome de 17 souches par séquençageSingle-Molecule Real-Time a révélé l’existence de un à troisgènes codant des effecteurs TAL par souche, pour un total dequatre gènes tal différents dont deux (tal23A et tal18H) ontété également transférés horizontalement entre Xcf et Xpp.L’ensemble de ces gènes constitue un répertoire spécifi que deXcf et Xpp qui pourrait expliquer la convergence pathologiqueentre ces pathovars. Des tests de pouvoir pathogène couplés àdes analyses transcriptomiques après inoculation d’un mutantde délétion de tal18H sur haricot ont révélé que TAL18H étaitimpliqué dans l’aggravation des symptômes et avait un effetpléiotrope sur le transcriptome du haricot lors de l’interaction.Les résultats de cette thèse constituent une
Common bacterial blight is the main bacterial disease of commonbean. This disease is caused by Xanthomonas citri pv.fuscans (Xcf) and X. phaseoli pv. phaseoli (Xpp). Xcf and Xppare phylogenetically distant yet they share the ability to inducethe same symptoms on common bean, which is suggestive ofpathological convergence between these two pathovars. Thisthesis aimed at identifying candidate genes for host specifi -city and studying the role of Transcription Activator-Like (TAL)effectors in the Xanthomonas – common bean interaction.Using a comparative genomic approach, we identifi ed 116genes specifi c to common bacterial blight agents, a largenumber of which were horizontally transferred between Xcfand Xpp. These genes encoded proteins involved in the differentsteps of the interaction.Single-Molecule Real-Timesequencing of 17 Xcf and Xpp genomes unveiled one to threeTAL-encoding genes per strain for a total of four different talgenes, two of which (tal23A and tal18H) were also horizontallytransferred between Xcf and Xpp. All these genes forma repertoire specifi c to Xcf and Xpp that could be responsiblefor the pathological convergence observed between thesetwo pathovars. Combination of pathogenicity tests and transcriptomicsafter inoculation of a tal18H deletion mutant oncommon bean plants revealed that TAL18H was involved insymptom development and displayed a pleiotropic effect oncommon bean transcriptome during the interaction. The resultsof this thesis constitute a stepping stone towards optimizingthe monitoring of co

APA, Harvard, Vancouver, ISO und andere Zitierweisen

41

Veysset, Jérémy. „Simulation des grands espaces et des temps longs“. Thesis, Paris, ENMP, 2014. http://www.theses.fr/2014ENMP0083/document.

Der volle Inhalt der Quelle

Annotation:

L'interaction fluide structure est présente dans beaucoup de problèmes industriels, dans les domaines d'ingénierie mécanique, civile ou biomécanique. Même si les performances informatiques s'améliorent considérablement et que les méthodes en mécanique numérique gagnent en maturité, certaines difficultés ne permettent pas encore de réaliser des simulations numériques précises. Actuellement deux méthodes numériques gagnent en popularité pour la simulation numérique d'interactions fluide structure: la méthode de partitionnement et la méthode monolithique. Des résultats de la littérature montrent que la première est efficace et précise mais qu'elle peut rencontrer des problèmes d'instabilité si les ratios de densité sont élevés ou que les géométries sont complexes. Les méthodes d'immersion sont de plus en plus utilisées par la communauté scientifique. Différentes approches ont été développées, dont la Méthode d'Immersion de Volume. Cette méthode permet de faciliter la mise en place des calculs. Ainsi il n'est pas nécessaire de construire des maillages concordant avec la géométrie des objets, et le couplage entre les fluides et les solides se fait naturellement. C'est sur cette analyse qu'a été développé le logiciel Thost. Il permet de simuler des procédés industriels tels que le chauffage de pièces métalliques dans les fours industriels ou la trempe sans caractériser expérimentalement des coefficients de transfert. Le but d'un tel logiciel est de permettre une meilleure compréhension des procédés et ainsi de les optimiser. Cependant les coûts de calcul restant élevés, le but de la thèse est de les diminuer en s'appuyant sur des méthodes numériques innovantes tels que l'adaptation dynamique de maillage anisotrope, des méthodes éléments finis stabilisées ou l'immersion directe des objets à partir de la Conception Assistée par Ordinateur
Fluid-Structure Interaction (FSI) describes a wide variety of industrial problems arising in mechanical engineering, civil engineering and biomechanics. In spite of the available computer performance and the actual maturity of computational fluid dynamics and computational structural dynamics, several key issues still prevent accurate FSI simulations.Two main approaches for the simulation of FSI problems are still gaining attention lately: partitioned and monolithic approaches. Results in the literature show that the partitioned approach is accurate and efficient but some instabilities may occur depending on the ratio of the densities and the complexity of the geometry. Monolithic methods are still of interest due to their capability to treat the interaction of the fluid and the structure using a unified formulation. In fact it makes the build up of a FSI problem easier as the mesh do not have to fit the geometry of the solids and the transfers are treated naturally.The software Thost has been created based on these analyzes. Thost is a 3D aerothermal numerical software. It has been developped for the numerical simulation of industrial processes like the heating in industrial furnaces as well as quenching. Its target is to model numericaly the thermal history of the industrial pieces in their environment without using any transfer coefficient. However the computational costs are still high and therefore the software is not fully efficient from an industrial point of view to simulate, analize and improve complex processes. All the work in this PhD thesis has been done to reduce the computational costs and optimize the accuracy of the simulations in Thost based on innovatives numerical methods such as dynamic anisotropic mesh adaptation, stabilized finite elements methods and immersing the objects directly from their Computer Aided Design files

APA, Harvard, Vancouver, ISO und andere Zitierweisen

42

Pensieri, Pasquale. „Rôle d'Otx2 dans les photorécepteurs de la rétine mature“. Electronic Thesis or Diss., Université Côte d'Azur (ComUE), 2019. http://www.theses.fr/2019AZUR6038.

Der volle Inhalt der Quelle

Annotation:

Si le rôle du facteur de transcription Otx2 dans le développement de la rétine est bien compris, sa fonction dans la rétine adulte reste floue. L'expression d’Otx2 est maintenue dans l'épithélium pigmenté rétinien (RPE), les photorécepteurs (PR) et les cellules bipolaires durant toute la vie. Des travaux antérieurs ont montré que l’ablation du gène Otx2 dans la rétine adulte conduit à une dégénérescence des PR, suggérant qu’il a un rôle dans leur maintenance. Il a ensuite été montré que l’expression d’Otx2 restreinte dans le RPE est à la fois nécessaire et suffisante pour la survie des photorécepteurs, ce qui indique une fonction neuroprotectrice non autonome dans ces cellules. Ces résultats n’expliquent pas la fonction de la protéine Otx2 endogène dans les PR. Afin de l'élucider, nous avons utilisé la lignée de souris Crx-CreERT2 inductible pour procéder à la délétion du gène Otx2 spécifiquement dans les PR. L'histologie a confirmé que l’expression d’Otx2 endogène n'est pas nécessaire à la survie des PR ni au maintien de leur identité cellulaire chez l'adulte. Cependant, nous avons constaté qu’Otx2 est impliqué dans l'adaptation visuelle en régulant le mouvement de l'arrestine induit par la lumière, dans les photorécepteurs. L’arrestine est transportée entre les segments externes du PR, où il s'accumule à la lumière, et les segments internes et le soma, où il s'accumule dans l'obscurité, pour moduler la sensibilité du PR, en réponse à différentes intensités lumineuses. Ce trafic est compromis après délétion d’Otx2 dans les PR : l’arrestine reste bloquée dans les segments internes et dans le soma, même après une exposition prolongée à la lumière. Des tests comportementaux ont montré que les souris mutantes sont fortement photophobes. Les mécanismes détaillés restent à élucider.Pour déchiffrer le réseau de gènes contrôlé par Otx2 dans les PR, nous avons effectué des analyses temporelles par RNA-seq au décours de l'inactivation d’Otx2 dans les PR. Aucun gène de la cascade de phototransduction ne semble dérégulé. Par contre, l’expression de certains gènes de la matrice extracellulaire et, de manière surprenante, d'un groupe de gènes spécifiquement impliqués dans la mélanogénèse dans le RPE, et déjà connus pour être des cibles directes d'Otx2, est diminuée. Un examen attentif des souris porteuses d’une délétion d’Otx2 spécifique dans les PR a confirmé que le niveau de protéine Otx2 dans les noyaux du RPE est réduit, tandis que la protéine Otx2 peut maintenant être détectée dans les segments externes de PR. Au vu de ces données et de l'effet neuro-protecteur d'Otx2 joué par le RPE, nous avons émis l'hypothèse d'un transfert direct de la protéine Otx2 du RPE aux PR. Pour prouver cette hypothèse, des vecteurs viraux dirigeant l'expression d’une forme d’Otx2 étiquetée, spécifiquement dans les cellules du RPE, ont été générés et utilisés pour l'injection sous-rétinienne chez la souris. Après transduction du RPE, nous avons recherché si la forme d’Otx2 étiquetée pouvait être détectée dans les PR. Nous avons constaté que dans la rétine contrôle, un transfert constitutif d'Otx2 est actif à faible débit. Après l'induction du KO dans les PR, le taux de ce transfert est augmenté, agissant probablement comme une réponse neuro-protectrice. Une telle augmentation du transfert explique bien la réduction du taux de protéine Otx2 dans le RPE et la diminution d’expression des gènes de la mélanogenèse. La protéine Otx2 transférée semble être transportée des segments externe et interne des PR vers les synapses situées dans la couche plexiforme externe, ce qui suggère que sa fonction neuro-protectrice met en jeu des mécanismes différents de sa fonction génomique classique de facteur de transcription.En conclusion, cette étude a révélé une nouvelle fonction de la protéine Otx2 endogène dans l'adaptation des PR à la lumière et a démontré l'existence d'un transfert de protéine Otx2 exogène, du RPE vers les PR, avec un rôle neuro-protecteur présumé
Although the role of Otx2 transcription factor in retinal development is well understood, its function in the adult retina remains unclear. Otx2 expression is maintained in Retinal Pigmented Epithelium (RPE), photoreceptors (PRs) and bipolar cells throughout life. Previous works from our lab showed that Otx2 ablation in the adult retina leads to photoreceptor degeneration, suggesting a role in their maintenance. It was then showed that RPE-restricted Otx2 expression was both necessary and sufficient for photoreceptor maintenance, indicating a non-cell-autonomous neuroprotective function. This left unexplained the function of endogenous Otx2 in PRs. In order to elucidate it, we performed PR-specific Otx2 knockout, using the inducible PR-specific Crx-CreERT2 mouse line. Histology confirmed that endogenous Otx2 is not required for PR survival or maintenance of their cell identity in the adult. However, we found that Otx2 is implicated in visual adaptation, by regulating the light induced movement of arrestin in photoreceptors. Arrestin shuttles between PR outer segments, where it accumulates in the light, and inner segments and soma, where it accumulates in the dark, to modulate PR sensitivity, in response to different light intensities. Such trafficking is compromised in PR-specific Otx2 KO mice, where arrestin remains blocked in inner segments and soma, even after prolonged light exposure. Behavioural tests showed that PR-specific Otx2 KO mice are strongly photophobic. The detailed mechanisms remain to be elucidated.To decipher the genetic network controlled by Otx2 in PRs, we carried out time series RNA-seq analyses following PR-specific Otx2 knockout. No photo-transduction gene was found to be deregulated. Instead of that, we found down-regulation of extracellular matrix genes and, surprisingly, of a group of RPE-specific genes involved in melanogenesis, genes that were already known to be direct targets of Otx2. Careful examination of PR-specific Otx2 KO mice confirmed that Otx2 protein level in the RPE nuclei was reduced while some Otx2protein could now be detected in PR outer segments. According to such data and to the neuroprotective effect of RPE-restricted Otx2, we hypothesized a direct Otx2 protein transfer from RPE to PRs. To prove this hypothesis, viral vectors driving tagged-Otx2 expression specifically in RPE cells were generated and used for sub-retinal injection in mice. After RPE transduction, tagged-Otx2 could be detected in the PRs. We found that in wild type retina, constitutive Otx2 transfer is active at low rate. After KO induction in the PRs, the transfer rate is increased, probably acting as a neuroprotective response. Such increment in the transfer explains well the reduction of Otx2 protein level in the RPE and the subsequent downregulation of melanogenesis genes. Transferred Otx2 appears to travel across PR cells from the inner and outer segments to the synapse in the outer plexiform layer, suggesting its neuroprotective function involves mechanisms that are different from its classical genomic function of transcription factor.In conclusion, this study disclosed a new function of endogenous Otx2 in PR light adaptation and demonstrated the existence of a transfer of Otx2 from RPE to PR cells, with a putative neuroprotective role

APA, Harvard, Vancouver, ISO und andere Zitierweisen

43

Mitchell, Jillian Mary Graham, und jill mitchell@health sa gov au. „A Matter of Urgency! Remote Aboriginal Womens Health. Examining the transfer, adaptation and implementation of an established holistic Aboriginal Well Womens Health program from one remote community to another with similar needs and characteristics“. Flinders University. Nursing and Midwifery, 2007. http://catalogue.flinders.edu.au./local/adt/public/adt-SFU20070725.112610.

Der volle Inhalt der Quelle

Annotation:

Aim: As a priority for Aboriginal women, in the context of worsening Aboriginal health and lack of clarity about successful strategies to address healthcare needs, this research explored successful strategies in remote Aboriginal womens health that may be transferable to another community with similar health needs. Methodology: Against a background of cultural and historical events, the study sought to identify existing strategies and frameworks for Aboriginal womens health. It uses Naturalistic Inquiry situated within the Interpretive paradigm and conceptualised within the philosophical approach of feminist and critical social theory It has examined Aboriginal health providers and womens priorities, practices, perceptions and expectations within the context of primary health care and community development principles by Participatory Action Research (PAR). The successful elements of an established and effective Aboriginal Well Womens Health (AWWH) program from Central Australia (CA) were identified, transferred and adapted to meet the needs of a willing recipient remote community in South Australia (SA). Working together with healthcare providers from CA and SA, the adapted Well Womens Health program was implemented in an Aboriginal Community Controlled Health Service collaboratively with local mainstream Community Womens health services and evaluated. Results: Over a two year period, the research was evaluated through Critical Social Theory examining both the process of implementation and the impact on the Aboriginal community, analysing both qualitative and quantitative data. The AWWH program model and its principles were successfully transferred, adapted and implemented in this community. The AWWH program which included comprehensive health screening, health information and lifestyle sessions have become core business of the Aboriginal health service and an Aboriginal Mens Well Health program has also been established using the same model. The women have found the AWWH program culturally acceptable and their attendance has steadily increased and the program has reached those women in the community who previously had never experienced a well health check. It has also identified an extremely high incidence and comorbidity of acute illness and chronic disease in diabetes, renal and dental disease, mental and social health problems that require address. Conclusion: Health programs that are well established and effective can be successfully replicated, transferred and adapted to other communities if the elements that made them successful are acknowledged and those principles are then transferred with the program to a willing community with similar needs. This program transfer has potential to save much time and developmental costs that will help to address poor Aboriginal health.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

44

Van, den Broek d'Obrenan Ghislain. „Adaptation du procédé RTM (Moulage par Transfert de Résine) à la mise en œuvre de matériaux composites à matrice thermoplastique“. Thesis, Lyon, INSA, 2011. http://www.theses.fr/2011ISAL0112/document.

Der volle Inhalt der Quelle

Annotation:

Le procédé « Resin Transfer Molding » (RTM) est très largement utilisé pour la production industrielle de matériaux composites à matrice thermodurcissable. En effet, de nombreux domaines tels que l’automobile et l’aéronautique l’emploi couramment. Dans ce travail nous avons adapté ce procédé à la mise en œuvre de matériaux composites à matrice thermoplastique afin de répondre aux critères écologiques et économiques imposés aux industries. Pour cela plusieurs étapes ont été nécessaires. La première fut la sélection d’une chimie robuste, adaptée aux exigences du procédé (faible viscosité initiale du système réactif, temps de polymérisation court, etc). La chimie choisie fut la polymérisation anionique par ouverture de cycle de l’ε-caprolactame dans le but d’obtenir du polyamide-6 (PA-6). Une étude rhéo-cinétique ainsi que les caractérisations physico-chimiques d’un PA-6 obtenu au laboratoire furent réalisé. A la suite de cette étape, des essais en conditions de procédé ont été effectué avec l’utilisation d’un équipement pilote dédié. Ces essais furent la source de modifications et d’optimisations de certains paramètres du procédé. La troisième étape, a consisté à la production de pièces composites avec un renfort de type : tissu unidirectionnel de verre. Cette production fut suivie de tests mécaniques et physico-chimiques afin d’évaluer les propriétés de ces pièces. Différents ensimages de tissu ont été étudiés avec, pour objectif, la détermination de celui offrant les meilleures propriétés. Durant cette étude nous avons observé que la nature de l’ensimage impactait peu la chimie. Pour finir, nous avons mis en place un ensimage réactif qui permettra une meilleure interaction fibre/matrice
The "Resin Transfer Molding" (RTM) process is very largely used for the industrial production of composites materials with thermoset matrix. Indeed, it’s used by many fields such as the automotive and aeronautics. In this work we adapted this process to the manufacture of composite materials with thermoplastic matrix in order to answer the ecological and economic criteria imposed on industries. For that several steps were necessary. The first was the selection of a robust chemistry, adapted to the requirements of the process (low initial viscosity of the reactive system, polymerization time, etc). The selected chemistry, was the ring opening polymerization of ε - caprolactam to obtain polyamide-6 (PA-6). Rhéo-kinetics studies, as well as the physicochemical characterizations of a Pa-6 obtained at the laboratory were carried out. Following this step, tests in conditions of process were carried out with the use of dedicated pilot equipment. These tests were the source of modifications and optimizations of certain parameters of the process. The third step, consisted with the production of composite parts with a reinforcement of the type: unidirectional glass fabric. This production was followed mechanical and physico-chemical tests in order to evaluate the properties of these parts. Various sizing of the glass fabric were studied with, for objective, to determine which to offer the best properties. During this study we observed the low impact of the sizing on the chemistry of PA-6. To finish, we set up a reactive sizing which will allow a better interaction fibre/matrix

APA, Harvard, Vancouver, ISO und andere Zitierweisen

45

Madera, Cristofer. „EXAMINATION OF THE EFFECT OF DIFFERENT TRAINING METRICS ON PERFORMANCE OF A MINIMALLY INVASIVE SURGERY TRANSFER TASK“. VCU Scholars Compass, 2013. http://scholarscompass.vcu.edu/etd/3276.

Der volle Inhalt der Quelle

Annotation:

The purpose of this experiment was to determine if there existed techniques to more efficiently train prospective surgeons the skills necessary to capably perform minimally invasive surgical procedures. Also, we wanted to know if trainees could be pushed to cognitively define a laparoscopic environment with a novel hand-eye relationship. To explore these questions, a simulation was setup wherein subjects would perform a laparoscopic transfer task and receive active feedback during training. Different subjects would receive different metrics as feedback and a comparison would be made between subjects with respect to standard metrics. Results of this experiment show that all subjects adapt to a laparoscopic environment and that they do so at different rates and to different proficiencies. The difference was shown to be statistically significant. It was concluded that the techniques we utilized were effective enough to claim as useful techniques to utilize in current training systems.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

46

Laurentis, Guilherme Lucas de. „Modelo de transferência de riscos hidrológicos como estratégia de adaptação às mudanças globais segundo cenários de vulnerabilidade dos recursos hídricos“. Universidade de São Paulo, 2012. http://www.teses.usp.br/teses/disponiveis/18/18138/tde-08012013-090756/.

Der volle Inhalt der Quelle

Annotation:

Esta pesquisa é parte do projeto FAPESP-IAV \"Assessment of Impacts and Vulnerability to Climate Change in Brazil and Strategies for Adaptation Options\". Avaliou-se a acoplagem entre o modelo climático regional Eta-CPTEC, o modelo hidrológico de grandes bacias MGB-IPH e o modelo de transferências de riscos hidrológicos MTRH-SHS como estratégia de adaptação à escassez hídrica em regimes fluviométricos sob mudanças de longo prazo. A metodologia foi aplicada em postos de monitoramento fluviométrico cujas áreas de drenagem variaram entre 386,6 e 10.929,9 \'KM POT.2\', pertencentes à bacia hidrográfica do Rio Piracicaba, com área total de 12.589 \'KM POT.2\'. Buscou-se avaliar a aptidão do MGB-IPH em representar as vazões mínimas médias anuais de sete dias consecutivos, Q7mín. A calibração e a validação do MGB-IPH corresponderam ao período 1971-1990. Com o MGB-IPH calibrado e validado, foram acoplados dados climáticos diários de saída do Eta-CPTEC, referentes ao período 2013-2099. Foram obtidas séries de Q7mín anuais que representaram a vulnerabilidade de escassez hídrica referentes a quatro cenários de mudanças globais para o período 2013-2099. Um quinto cenário, sem mudanças, foi proposto como a manutenção de características de Q7mín anuais observados. Foram gerados cenários para cinco sub-bacias, com áreas de 927,1 a 10.929,9 \'KM POT.2\'. Procedeu-se à análise de riscos, associando os prejuízos econômicos dos impactos decorrentes das vazões mínimas anuais nos cinco cenários propostos. Para o MTRH-SHS selecionaram-se a série de mínimos anuais, Q7mín-ano, em cada cenário, associado a um prejuízo distribuído pela área da bacia. Para o ressarcimento dos prejuízos, otimizou-se o prêmio a ser pago pela aplicação do MTRH-SHS sob diferentes limites de cobertura do fundo de seguros, curvas de danos regionais e para tempos de retorno de 10 até 100 anos. Os resultados demonstraram diferentes impactos das mudanças globais conforme cada cenário. A variação climática resultou em impacto de +24,8% na Q7mín média futura. Por sua vez, a expansão da urbanização dos municípios da bacia representou impacto de +0,2% na variação da Q7mín. A operação do Sistema Cantareira, que atua nas cabeceiras, potencialmente interfere em -14,6% da disponibilidade hídrica da bacia. As combinações de mudanças dos cenários resultaram em prejuízos e prêmios de acordo com a cobertura selecionada. Os prejuízos foram de 9,5 a 1.437 R$/\'KM POT.2\' e os prêmios otimizados para ressarcir esses prejuízos situaram-se entre 9,9 e 1.479 R$/\'KM POT.2\'. Concluiu-se sobre a viabilidade da estratégia de acoplagem entre os modelos como alternativa de adaptação para o planejamento estratégico na recuperação de bacias hidrográficas vulneráveis a eventos extremos de escassez hídrica.
This research is part of the project FAPESP-IAV \"Assessment of Impacts and Vulnerability to Climate Change in Brazil and Strategies for Adaptation Options\". It was evaluated the coupling among the regional climate model Eta-CPTEC, the hydrological model for large basins MGB-IPH and the hydrologic risks transfer model MTRH-SHS as an adaptation strategy to cope with water scarcity in flow regimes under long-term changes. The methodology was applied to fluviometric monitoring stations whose drainage areas varied between 386.6 and 10,929.9 \'KM POT.2\', belonging to the Piracicaba River basin with total area of 12.589 \'KM POT.2\'. The focus was on improving the performance of the MGB-IPH in representing the average of the annual seven day minimum flows, Q7mín. The calibration and validation of the MGB-IPH corresponded to the period 1971-1990. With the MGB-IPH calibrated and validated, the Eta-CPTEC output daily climate data were coupled, related to the 2013-2099 period. It was obtained sets of annual Q7mín that represented the vulnerability to water scarcity related to four scenarios of global change for the 2013-1099 period. A fifth scenario, without changes, was proposed as the maintenance of the characteristics of annual Q7mín observed. Scenarios were generated for five sub-basins with areas ranging from 927.1 to 10,929.9 \'KM POT.2\'. The risk analysis was done, relating the impacts of economic losses due to the occurrence of the annual minimum flows over the five proposed scenarios. To the MTRH-SHS a set of annual minima (Q7mín-year) was selected in each scenario, associated to the losses distributed along the watershed area. For the losses compensation, the premium to be paid was optimized through the implementation of the insurance fund, under different coverage limits and regional damage curves, for return periods from 10 to 100 years. The results showed different impacts of global changes in each scenario. The climate change resulted in an impact of +24.8% on the Q7mín future average. In turn, the expansion of urbanization of the watershed municipalities represented an impact of +0.2% on the variation of the Q7mín. The operation of the Cantareira System, which operates in the headwaters, potentially interferes with -14.6% on the water availability in the basin. The combinations of the scenarios of changes resulted in losses and premiums depending on the coverage selected. The damages ranged from 9.5 to 1.437 R$/\'KM POT.2\' while the optimized premiums to compensate these losses varied between 9.9 and 1479 R$/\'KM POT.2\'. It was concluded that is feasible this strategy of model coupling as an adaptation alternative to strategic planning in the recovery of watersheds vulnerable to extreme water shortage.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

47

Dragović, Njegoš, und Snežana Urošević. „Using a renewable energy resources in the clothing industry“. Thesis, Київський національний університет технологій та дизайну, 2021. https://er.knutd.edu.ua/handle/123456789/19082.

Der volle Inhalt der Quelle

Annotation:

The clothing industry is one of the most affected branches of the economy in crisis conditions such as the global Covid-19. If the production of textiles took place in the existing conditions, many companies would stop because the labor costs are high. From the processing of raw materials, through the production of textiles, all the way to the production of clothing, the greatest chance to return to real trends is the energy transition, technological transfer and adjustment of workers. The aim of this paper is to point out the possibilities that would keep the textile industry in line with global trends in the use of renewable energy sources.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

48

Lévy-Bencheton, Delphine. „Perception visuelle et plasticité oculomotrice : aspects fondamentaux et application clinique“. Phd thesis, Université Claude Bernard - Lyon I, 2013. http://tel.archives-ouvertes.fr/tel-00926062.

Der volle Inhalt der Quelle

Annotation:

Une façon d'explorer visuellement notre monde consiste à déplacer très rapidement nos yeux pour en analyser le contenu. Dans certaines circonstances, ces mouvements oculaires, appelés saccades, peuvent perdre de leur précision. Fort heureusement notre cerveau est capable de corriger cette imprécision en ajustant progressivement la taille de ces saccades grâce à des mécanismes de plasticité cérébrale : c'est l'adaptation saccadique. L'adaptation saccadique est souvent utilisée comme modèle d'étude des mécanismes de plasticité visuo-motrice. Nous faisons également l'hypothèse qu'elle puisse servir d'outil thérapeutique. Dans cette thèse, nous nous sommes essentiellement intéressés à l'adaptation des saccades volontaires en augmentation d'amplitude chez des sujets contrôles et chez des patients souffrant d'une amputation homonyme du champ visuel (hémianopsie latérale homonyme). Les expériences 1 à 3, réalisées chez le sujet contrôle, ont permis de découvrir les repères utilisés par le cerveau pour réaliser cette adaptation sensori-motrice et coder l'environnement visuel (expérience 1), et les mécanismes d'adaptation saccadique quand la cible visuelle n'est pas directement codée (remapping dans l'expérience 2 et cible virtuelle dans l'expérience 3). Enfin l'expérience 4 propose une application clinique du protocole d'adaptation des anti-saccades utilisé lors de l'expérience 3, dans un but thérapeutique de rééducation comportementale de patients hémianopsiques.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

49

Das, Manirupa. „Neural Methods Towards Concept Discovery from Text via Knowledge Transfer“. The Ohio State University, 2019. http://rave.ohiolink.edu/etdc/view?acc_num=osu1572387318988274.

Der volle Inhalt der Quelle

APA, Harvard, Vancouver, ISO und andere Zitierweisen

50

Knöppel, Anna. „Experimental Evolution : and Fitness Effects of Mutations“. Doctoral thesis, Uppsala universitet, Institutionen för medicinsk biokemi och mikrobiologi, 2016. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-306592.

Der volle Inhalt der Quelle

Annotation:

Bacteria have small, streamlined genomes and evolve rapidly. Their large population sizes allow selection to be the main driver of evolution. With advances in sequencing technologies and precise methods for genetic engineering, many bacteria are excellent models for studying elementary questions in evolutionary biology. The work in this thesis has broadly been devoted to adaptive evolution and fitness effects of different types of mutations. In Paper I we experimentally tested the fitness constrains of horizontal gene transfer (HGT), which could be used to predict how the fixation of HGT events are affected by selection and fitness effects. We found that the majority of the examined HGT inserts were indistinguishable from neutral, implying that extra DNA transferred by HGT, even though it does not confer an immediate selective advantage, could be maintained at transfer-selection balance and serve as a reservoir for the evolution of novel beneficial functions. Paper II examined why four synonymous mutations in rpsT (encoding ribosomal protein S20) reduced fitness, and how this cost could be genetically compensated. We found that the cause for the fitness reduction was low S20 levels and that this lead to a defective subpopulation of 30S subunits lacking S20. In an adaptive evolution experiment, these impairments were compensated by up-regulation of S20 though various types of mutations. In Paper III we continued the studies of how the deleterious rpsT mutations could be compensated. The mutations either down-regulated the global regulator Fis or altered a subunit of the RNA polymerase (rpoA). We found that the decreased S20 levels in the cells causes an assembly defect of the 30S particles and that the fis and rpoA mutations restored the skewed S20:ribosome ratio by both increasing S20 levels and decreasing other ribosomal components. Paper IV examined adaptation of two bacterial species to different growth media. A total of 142 different adaptive mutations were identified and 112 mutants were characterized in terms of fitness. We found that the experimental variation in fitness measurements could be reduced 10-fold by introducing some adaptive mutations prior to the experiment, allowing measurements of fitness differences as small as 0.04%.

APA, Harvard, Vancouver, ISO und andere Zitierweisen

Dissertationen zum Thema „Transfer of adaptation“

Geben Sie eine Quelle nach APA, MLA, Chicago, Harvard und anderen Zitierweisen an