Дисертації з теми "Distribution Shift"
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся з топ-50 дисертацій для дослідження на тему "Distribution Shift".
Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.
Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.
Переглядайте дисертації для різних дисциплін та оформлюйте правильно вашу бібліографію.
Zennaro, Fabio. "Feature distribution learning for covariate shift adaptation using sparse filtering." Thesis, University of Manchester, 2017. https://www.research.manchester.ac.uk/portal/en/theses/feature-distribution-learning-for-covariate-shift-adaptation-using-sparse-filtering(67989db2-b8a0-4fac-8832-f611e9236ed5).html.
Повний текст джерелаIngram, Douglas Russell. "Starlight correlated with damped Lyman alpha absorbers /." Thesis, Connect to this title online; UW restricted, 1996. http://hdl.handle.net/1773/5435.
Повний текст джерелаNodet, Pierre. "Biquality learning : from weakly supervised learning to distribution shifts." Electronic Thesis or Diss., université Paris-Saclay, 2023. http://www.theses.fr/2023UPASG030.
Повний текст джерелаThe field of Learning with weak supervision is called Weakly Supervised Learning and aggregates a variety of situations where the collected ground truth is imperfect. The collected labels may suffer from bad quality, non-adaptability, or insufficient quantity. In this report, we propose a novel taxonomy of Weakly Supervised Learning as a continuous cube called the Weak Supervision Cube that encompasses all of the weaknesses of supervision. To design algorithms capable of handling any weak supervisions, we suppose the availability of a small trusted dataset, without bias and corruption, in addition to the potentially corrupted dataset. The trusted dataset allows the definition of a generic learning framework named Biquality Learning. We review the state-of-the-art of these algorithms that assumed the availability of a small trusted dataset. Under this framework, we propose an algorithm based on Importance Reweighting for Biquality Learning (IRBL). This classifier-agnostic approach is based on the empirical estimation of the Radon-Nikodym derivative (RND), to build a risk-consistent estimator on reweighted untrusted data. Then we extend the proposed framework to dataset shifts. Dataset shifts happen when the data distribution observed at training time is different from what is expected from the data distribution at testing time. So we propose an improved version of IRBL named IRBL2, capable of handling such dataset shifts. Additionally, we propose another algorithm named KPDR based on the same theory but focused on covariate shift instead of the label noise formulation. To diffuse and democratize the Biquality Learning Framework, we release an open-source Python library à la Scikit-Learn for Biquality Learning named biquality-learn
Chen, Xiaojun. "Multiple Scattering from Bubble Clouds." Scholarly Repository, 2010. http://scholarlyrepository.miami.edu/oa_theses/36.
Повний текст джерелаHo, Yo-Bin. "Modeling the distribution and shift of Chinese Americans in the United States : an analysis by trend surface expansion /." The Ohio State University, 1995. http://rave.ohiolink.edu/etdc/view?acc_num=osu1487861796820923.
Повний текст джерелаBrodén, Daniel. "Analysis of Demand Response Solutions for Congestion Management in Distribution Networks." Thesis, KTH, Industriella informations- och styrsystem, 2013. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-138575.
Повний текст джерелаFigueiredo, Cláudia Cerqueira de Souza Machado Vieira. "Repartição funcional do rendimento no Brasil : uma abordagem setorial." Master's thesis, Instituto Superior de Economia e Gestão, 2019. http://hdl.handle.net/10400.5/19042.
Повний текст джерелаNeste trabalho faz-se uma análise da evolução da repartição funcional do rendimento no período contemporâneo Brasileiro, entre 2000 e 2016. A alteração da parte do trabalho no rendimento gerado na economia (VAB ou PIB), o chamado labour share, resulta da combinação de duas tendências importantes, a saber, a evolução da produtividade do trabalho e a do salário médio. Por sua vez, a alteração global (ou média) destas variáveis resulta das tendências verificadas em cada setor da economia em causa, cada um com as suas peculiaridades. Depois de um breve enquadramento teórico e revisão da literatura relevante, para a análise empírica da evolução do caso brasileiro será dada atenção à imputação do rendimento misto a "salários" e "lucros". Com isto pretende-se observar as relações existentes entre a evolução da participação do trabalho na economia e as transformações registadas na estrutura produtiva. Faz-se ainda a aplicação de uma análise shift-share, através da qual é possível distinguir/quantificar o chamado efeito "intra-sectorial" ou "within" (evolução do labour share de cada setor) do efeito "inter-setorial" ou "between" (ou de mudança estrutural, i.e., alterações no peso relativo de cada setor).
This paper analyzes the evolution of the functional distribution of income in the contemporary Brazilian period between 2000 and 2016. The change in the share of labor in the income generated in the economy (GVA or GDP), the so-called labor share, results from the combination of two important trends, namely the evolution of labor productivity and that of the average wage. In turn, the overall (or average) change in these variables results from trends in each sector of the economy concerned, each with its own peculiarities. After a brief theoretical framework and review of the relevant literature, for the empirical analysis of the evolution of the Brazilian case, attention will be given to the imputation of mixed income to "wages" and "profits". With this it is intended to observe the existing relations between the evolution of the participation of the work in the economy and the transformations registered in the productive structure. A shift-share analysis is also applied, whereby it is possible to distinguish / quantify the so-called "intra-sectoral" or "within" effect (evolution of each sector's labor share) from the "inter-sectoral" or "inter-sectoral" effect. "Between" (or structural change, ie, changes in the relative weight of each sector).
info:eu-repo/semantics/publishedVersion
Ryckebusch, Gaëlle. "Analysis of Demand-Response Participation Strategies for Congestion Management in an Island Distribution Network." Thesis, KTH, Industriella informations- och styrsystem, 2015. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-169220.
Повний текст джерелаSlove, Davidson Jessica. "The plasticity and geography of host use and the diversification of butterflies." Doctoral thesis, Stockholms universitet, Zoologiska institutionen, 2012. http://urn.kb.se/resolve?urn=urn:nbn:se:su:diva-69573.
Повний текст джерелаAt the time of the doctoral defence,the following papers were unpublished and had a status as follows: Paper 3: Submitted; Papers 4 and 5: Manuscripts
Fisco, Dana. "Reef Fish Spatial Distribution and Benthic Habitat Associations on the Southeast Florida Reef Tract." NSUWorks, 2016. http://nsuworks.nova.edu/occ_stuetd/408.
Повний текст джерелаAgnolini, Sébastien. "Contributions à l'étude et à la réalisation d'un système de distribution quantique de clef par codage en phase." Paris 6, 2007. https://pastel.archives-ouvertes.fr/pastel-00003416.
Повний текст джерелаDadalto, Câmara Gomes Eduardo. "Improving artificial intelligence reliability through out-of-distribution and misclassification detection." Electronic Thesis or Diss., université Paris-Saclay, 2024. http://www.theses.fr/2024UPASG018.
Повний текст джерелаThis thesis explores the intersection of machine learning (ML) and safety, aiming to address challenges associated with the deployment of intelligent systems in real-world scenarios. Despite significant progress in ML, concerns related to privacy, fairness, and trustworthiness have emerged, prompting the need for enhancing the reliability of AI systems. The central focus of the thesis is to enable ML algorithms to detect deviations from normal behavior, thereby contributing to the overall safety of intelligent systems.The thesis begins by establishing the foundational concepts of out-of-distribution (OOD) detection and misclassification detection in Chapter 1, providing essential background literature and explaining key principles. The introduction emphasizes the importance of addressing issues related to unintended and harmful behavior in ML, particularly when AI systems produce unexpected outcomes due to various factors such as mismatches in data distributions.In Chapter 2, the thesis introduces a novel OOD detection method based on the Fisher-Rao geodesic distance between probability distributions. This approach unifies the formulation of detection scores for both network logits and feature spaces, contributing to improved robustness and reliability in identifying samples outside the training distribution.Chapter 3 presents an unsupervised OOD detection method that analyzes neural trajectories without requiring supervision or hyperparameter tuning. This method aims to identify atypical sample trajectories through various layers, enhancing the adaptability of ML models to diverse scenarios.Chapter 4 focuses on consolidating and enhancing OOD detection by combining multiple detectors effectively. It presents a universal method for ensembling existing detectors, transforming the problem into a multi-variate hypothesis test and leveraging meta-analysis tools. This approach improves data shift detection, making it a valuable tool for real-time model performance monitoring in dynamic and evolving environments.In Chapter 5, the thesis addresses misclassification detection and uncertainty estimation through a data-driven approach, introducing a practical closed-form solution. The method quantifies uncertainty relative to an observer, distinguishing between confident and uncertain predictions even in the face of challenging or unfamiliar data. This contributes to a more nuanced understanding of the model's confidence and helps flag predictions requiring human intervention.The thesis concludes by discussing future perspectives and directions for improving safety in ML and AI, emphasizing the ongoing evolution of AI systems towards greater transparency, robustness, and trustworthiness. The collective work presented in the thesis represents a significant step forward in advancing AI safety, contributing to the development of more reliable and trustworthy machine learning models that can operate effectively in diverse and dynamic real-world scenarios
Ji, Ming Chao. "Molecular relaxation dynamics of Anthracene cations studied in an electrostatic storage ring." Thesis, Lyon 1, 2015. http://www.theses.fr/2015LYO10051.
Повний текст джерелаThe polycyclic aromatic hydrocarbon (PAH) molecules have been considered as possible carrier of the unidentified infrared emission bands from the interstellar medium (ISM) for about thirty years. The cooling dynamics of the PAH molecules which is essential to estimate their photostability and therefore their lifetime and size distributions in the ISM, has attracted numerous theoretical and experimental studies. In recent years, electrostatic storage devices (ESD) became powerful tool to investigate the cooling regime of molecules and clusters in a large time range from microseconds to seconds. Generally speaking, the decay of the emitted neutral yields due to dissociation of molecular cations or electron detachment of anions in such experiments carries information on the internal energy of the stored molecular ions. In this thesis work, the cooling regimes of anthracene cations are studied by following the time evolution of the internal energy distribution (IED) of the stored anthracene cations. A spontaneous neutral yield curve obtained from the stored molecular ions as a function of the storage time shows three distinguishable regions. The three regions indicate different cooling regimes at corresponding storage time range, i.e., the dissociation mechanism of the molecule dominates at storage time t < 1 ms, quenching of the dissociation by radiative cooling processes occurs during 1 < t < 3 ms and radiative cooling governs at t > 3 ms
Kwan, Kermit S. Jr. "The Role of Penetrant Structure on the Transport and Mechanical Properties of a Thermoset Adhesive." Diss., Virginia Tech, 1998. http://hdl.handle.net/10919/30666.
Повний текст джерелаPh. D.
Bourgeaud, Luana. "Histoire évolutive et potentiel adaptatif : une approche éco-évolutive de la vulnérabilité des espèces aux changements climatiques." Thesis, Toulouse 3, 2020. http://www.theses.fr/2020TOU30263.
Повний текст джерелаUnderstanding the processes influencing the geographical distribution of species is one of the main aims in ecology and is of particular interest since climate change caused by human activities is currently leading to the geographical redistribution of species. In this context, we explored the temporal dynamics of the climatic niche (the set of climatic requirements of a species) in an attempt to determine species' ability to cope with climate change. To do this, we investigated historical rates of climatic niche change which describe past climatic niche changes that occurred over the course of species evolutionary history assuming that past climatic changes reflect current climate change. Following the reconstruction of a dated phylogeny for 12,616 fish species, we compiled climate and distribution data to estimate historical rates of climatic niche change. In a first project, we studied the factors influencing historical rates of climatic niche change in aquatic environments. In particular, we confirmed that historical rates of climatic niche change are greater at higher latitudes. We also discussed the differences between marine and freshwater ecosystems. In a second project, we correlated historical rates of climatic niche change with measures of contemporary range shifts. We showed that historical rates of climatic niche change are positively associated with contemporary range shifts in marine fish. Our results suggest that the influence of historical rates of climatic niche change and more generally of species evolutionary history on their response to current climate change needs to be further explored to determine how it can inform biodiversity conservation
Ousmen, Ahmad. "Détermination du sens clinique d'un changement pour les questionnaires de qualité de vie relative à la santé en cancérologie." Thesis, Bourgogne Franche-Comté, 2019. http://www.theses.fr/2019UBFCE001/document.
Повний текст джерелаIn oncology, the health-related quality of life (HRQOL) is generally considered as a second endpoint in the clinical trials. The interpretation of the results of the longitudinal analysis of such data must be made in both statistical and clinical point of view in order to produce meaningful results for both patients and clinicians. The main objective is to assess the impact of the treatment on patient’s HRQOL level over time. The minimal clinically important difference (MCID) was defined as the smallest change between two scores in a treatment outcome that a patient would identify as important. Indeed, the longitudinal analysis of HRQOL remains complex, particularly due to the potential occurrence of a Response Shift effect characterizing the process of adaptation of the patient in relation to the illness and its treatment. Hence, the first objective of this work is a literature review concerning the determination of the MCID by the most commonly used methods: anchor-based and distribution-based methods. Secondly, calculating the MCID using anchor-based and distribution-based methods by applying different distribution criteria and several different anchors. The objective is to compare the results obtained by the different methods and to compare them to others obtained by previous studies. Finally, studying the impact of the Response Shift effect on the determination of MCID for the HRQOL questionnaires in cancer research using several data corresponding to different therapeutic situations and cancer locations
Falck, Lina, and Moa Thörlind. "YouTube och Musikbranschen : En studie i hur artisters användande av YouTube som marknadsföringskanal ser ut i en allt mer digital musikbransch." Thesis, Linnéuniversitetet, Institutionen för marknadsföring (MF), 2015. http://urn.kb.se/resolve?urn=urn:nbn:se:lnu:diva-35919.
Повний текст джерелаThis paper aims to analyze how music artists use YouTube as a marketing tool, and the importance YouTube has for these artists' marketing. A qualitative and inductive approach was used, while a number of interviews and a content analysis conducted as the basis for the empirical material. This to get an insight into the music industry's views on the topic, and also give an objective picture of how the platform is used as a marketing tool. Finally, the study is meant to give artists and others working in the music industry a picture of how the use of this marketing channel can be improved to create a more effective marketing. The theoretical framework was based on earlier theories of YouTube's importance, roles and potential in music artists' marketing. Theories on how YouTube enables artists to be discovered by A & R people or new fans, YouTube's impact on the music video's role as a marketing medium, and theories about the platform's impact on the balance of power within the industry has been the foundatioin. Theories about a new focus on opinion leaders; whom and when, as well as theories on relationship marketing, communication, feedback and sampling have also been applied. Finally, theories of YouTube's negative aspects, such as difficulties to stand out and illegal distribution has been looked into. In the account of the empirical material responses from the interviews with music artists and key people in the industry are included, as well as the results from the content analysis conducted by the interviewed artists' respective YouTube channels. In the analysis the theoretical framework is linked to the empirical material in a discussion and analysis. This is to compare previous theories with the industry's current opinions on the research question and the performers' actual usage of the platform, also to illustrate how music artists could improve their marketing through YouTube for more successful marketing. The conclusions of the essay is of the opinion that YouTube has evolved into a powerful marketing channel for music artists. The platform is of great importance for music artists interms of opportunities of exposure, but additional marketing activities are however required gor the material to be found by A & R departments and new fans. The music video has a continued important role for artists' exposure opportunities. YouTube creates an opportunity to build relationships between artists and fans, but this opportunity is generally not utilized fully by the artists. Feedback via the comment field could be utilized for greater levels of learning, and for stronger relationship building. One negative aspect of YouTube is that it can be considered as a way to illegaly distribute content, and it does add to reduced control over the artists own material. The creative authority seems with YouTube have moved more towards the artists themselves rather than their labels or management, but networks and resources available to the major labels seem to remain important to provide opportunities to be able stand out in the masses f available media on YouTube.
Delava, Émilie. "Impacts du réchauffement climatique sur la distribution géographique des insectes et mise en place des adaptations locales : cas d'un parasitoïde de drosophiles dans le sud-est de la France." Thesis, Lyon 1, 2013. http://www.theses.fr/2013LYO10315.
Повний текст джерелаPredicting biodiversity responses to anthropogenic climate change has become a field of research with major scientific and societal issues. The main goal of my thesis was to evaluate the impacts of global warming on a Drosophila parasitoid, Leptopilina boulardi, at a small geographical scale, the South-East of France. The aim was not only to examine the change in the distribution of the parasitoid in response to rising temperatures, but also to understand the adaptations associated with this change. First, the analysis of insect sampling and meteorological data allowed me to demonstrate a rapid expansion of the parasitoid range to the north with an average rate of 90km/decade as well as a simultaneous temperature increase of 1.57°C on average over the past 30 years in the studied area. Following the identification of the main environmental factors structuring the spatial distribution of L. boulardi, I fitted a model predicting its potential distribution in the south-east of France, under the current climate and in 2050, for two CO2 emission scenarios. In 2050, the geographical distribution of L. boulardi should significantly extend northward as a result of climate change. Then, by measuring several life history traits under four fluctuating temperature regimes, I have shown that populations of L. boulardi located on the border of the range are genetically differentiated from those in the central range. The fact that marginal populations have a greater fitness at low temperature suggests local adaptation of parasitoids in the area of progression of range. The last part of this thesis aimed to better understand the process of colonization of L. boulardi. For this, I undertook the development of RAD-sequencing markers to genotype 15 populations of this species distributed along a cline of latitude in the southeast of France. Numerous data from Illumina sequencing will allow me to characterize the genetic structure of the populations. All the results obtained in my thesis highlight the force with which climate change may impact species, in particular those of high trophic level, causing rapid changes in distribution along with genotypic and phenotypic changes underlying local adaptation
Elkington, Ruari B. "The education market for documentary film: Digital shifts in an age of content abundance." Thesis, Queensland University of Technology, 2016. https://eprints.qut.edu.au/98494/4/Ruari_Elkington_Thesis.pdf.
Повний текст джерелаGuio, Patrick. "Etude de paramètres ionosphériques à l'aide des raies de plasma électroniques observées à Eiscat." Université Joseph Fourier (Grenoble), 1998. http://www.theses.fr/1998GRE10253.
Повний текст джерелаSteige, Kim A. "Cis-regulatory variation and divergence in Capsella." Doctoral thesis, Uppsala universitet, Evolutionsbiologi, 2016. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-268953.
Повний текст джерелаBodén, Andreas. "The anode and the electrolyte in the MCFC." Doctoral thesis, KTH, Kemiteknik, 2007. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-4382.
Повний текст джерелаEtt av den svenska regeringens mål är att öka användandet av förnyelsebara bränslen och bränslen från biomassa. Bränsleceller och framförallt MCFC är användbara för dessa typer av bränslen. Den svenska marknaden kan dra fördelar av MCFC på två sätt; ökad bränsleutnyttjandegrad och utnyttjande av producerad värme för fjärrvärme. De flesta kommersiella MCFC-systemen idag är optimerade för användning av metan. Möjligheten att använda biomassa på den svenska marknaden gör det viktigt att studera hur MCFC kan anpassas eller optimeras för bra prestanda och låg degradering för användning med gas från biomassa eller andra förnyelsebara bränslen. Fokus i denna avhandling är på metoder som kan användas för att undersöka och utvärdera MCFC-elektroder och -elektrolyter med förnyelsebara bränslen, dvs. gaser innehållande CO2. Metoderna och resultaten är både experimentella och matematiskt modellerade. Målet med denna avhandling är att bättre förstå hur anodens prestanda beror på användningen av olika bränslen. Anodens kinetik och vattengasskiftreaktionen har studerats liksom möjligheten att förlänga cellens livstid genom att öka den initiala mängden elektrolyt medelst användning av anoden som reservoar. Effekten av segregation av katjoner i elektrolyten under last har också undersökts. Om gassammansättningen är i jämvikt enligt vattengasskiftreaktionen vid inloppet till strömtilledaren kommer gassammansättningen att vara nära uniform inuti elektroden. Om ingående gas inte är i jämvikt kommer stora koncentrationsgradienter uppkomma i strömtilledaren och påverka gassammansättningen i elektroden. Omsättningen med avseende på vattenskiftreaktionen av gasen i flödeskanalen verkar vara beroende av gasens flödeshastighet. För en anod som används i en uppfuktad blandning av vätgas och koldioxid som inte är i jämvikt befanns det att Ni har en viss löslighet i (Li/Na)2CO3. För att kunna använda anoden som reservoar för elektrolyt för att förlänga livstiden för MCFC skall anodens porstorleksfördelning överensstämma med katodens och ha en bimodal porstorleksfördelning för att ge en tillräckligt god prestanda i ett så stort elektrolytfyllnadsgradsintervall som möjligt. Modelleringsresultat för segregering av katjoner i elektrolyten under drift visar att litiumjoner anrikas i anoden för båda typerna av elektrolyt som används i MCFC. Elektrolytkoncentrationsförändringarna är små men kan behövas tas i beaktande vid långa driftstider. Denna avhandlings resultat kan användas för att bättre förstå hur MCFC skall anpassas för drift med förnyelsebara bränslen och hur elektroder kan utformas för att förlänga livstiden.
QC 20100630
Gallo, Diego Sanchez. "Arquitetura de IPTV com suporte à apresentação deslocada no tempo baseada em distribuição peer-to-peer." Universidade de São Paulo, 2009. http://www.teses.usp.br/teses/disponiveis/3/3141/tde-18112009-092103/.
Повний текст джерелаTelecommunication companies are suffering from the increasing offer of cheap and reliable voice over IP services, being forced to diversify their services looking for new revenue possibilities. Since these companies have a vast infrastructure, they are now providing TV services through the same telephony and data infrastructure, using their IP networks to offer IPTV. The goal of the present work is to allow, in this scenario, that such companies offer, additionally to the traditional TV services (e.g., the linear transmissions of the TV channels), differentiated services through the same infrastructure. The focus of the present work is, therefore, the offering of the time-shift service, allowing users to watch linear transmitted contents, time-shifted, without the need for any in-advance configuration. This approach gives more flexibility to the users, allowing them to choose the most appropriate time to watch some content without having to specify their interests in advance (i.e., without configuring some equipment to record the content or knowing in advance which programs will interest themselves). To achieve this goal, technologies for content transmission and distribution, as well as the peer-to-peer paradigm for file sharing were studied, resulting in the development of an architecture capable of offering the traditional linear transmissions service as well as the possibility of time-shift, combining multicast, distributed caching and peer-to-peer technologies. Accordingly, an efficient solution was envisioned, making use of all available resources in the system, including idle resources in the user equipments, to help in the caching and distribution of the time-shifted contents. Finally, a prototype was developed as a proof-of-concept for the designed architecture, which together with the performed tests, shows the viability of utilizing P2P networks in the distribution of time-shifted contents.
Sirois-Delisle, Catherine. "Modeling Future Climate Change Impacts on North American Bumblebee Distributions." Thesis, Université d'Ottawa / University of Ottawa, 2017. http://hdl.handle.net/10393/37028.
Повний текст джерелаCox, Dennis R. "A business approach for detecting shifts in defect counts modeled by the Poisson distribution /." Available to subscribers only, 2006. http://proquest.umi.com/pqdweb?did=1136092151&sid=8&Fmt=2&clientId=1509&RQT=309&VName=PQD.
Повний текст джерелаRapacciuolo, Giovanni. "Predicting species' range shifts under global change : when can species distribution models be useful?" Thesis, Imperial College London, 2013. http://hdl.handle.net/10044/1/18025.
Повний текст джерелаVenne, Simon. "Can Species Distribution Models Predict Colonizations and Extinctions?" Thesis, Université d'Ottawa / University of Ottawa, 2018. http://hdl.handle.net/10393/38465.
Повний текст джерелаBickel, Steffen. "Learning under differing training and test distributions." Phd thesis, Universität Potsdam, 2008. http://opus.kobv.de/ubp/volltexte/2009/3333/.
Повний текст джерелаEines der wichtigsten Probleme im Maschinellen Lernen ist das Trainieren von Vorhersagemodellen aus Trainingsdaten und das Ableiten von Vorhersagen für Testdaten. Vorhersagemodelle basieren üblicherweise auf der Annahme, dass Trainingsdaten aus der gleichen Verteilung gezogen werden wie Testdaten. In der Praxis ist diese Annahme oft nicht erfüllt, zum Beispiel, wenn Trainingsdaten durch Fragebögen gesammelt werden. Hier steht meist nur eine verzerrte Zielpopulation zur Verfügung, denn Teile der Population können unterrepräsentiert sein, nicht erreichbar sein, oder ignorieren die Aufforderung zum Ausfüllen des Fragebogens. In vielen Anwendungen stehen nur sehr wenige Trainingsdaten aus der Testverteilung zur Verfügung, weil solche Daten teuer oder aufwändig zu sammeln sind. Daten aus alternativen Quellen, die aus ähnlichen Verteilungen gezogen werden, sind oft viel einfacher und günstiger zu beschaffen. Die vorliegende Arbeit beschäftigt sich mit dem Lernen von Vorhersagemodellen aus Trainingsdaten, deren Verteilung sich von der Testverteilung unterscheidet. Es werden verschiedene Problemstellungen behandelt, die von unterschiedlichen Annahmen über die Beziehung zwischen Trainings- und Testverteilung ausgehen. Darunter fallen auch Multi-Task-Lernen und Lernen unter Covariate Shift und Sample Selection Bias. Es werden mehrere neue Modelle hergeleitet, die direkt den Unterschied zwischen Trainings- und Testverteilung charakterisieren, ohne dass eine einzelne Schätzung der Verteilungen nötig ist. Zentrale Bestandteile der Modelle sind Gewichtungsfaktoren, mit denen die Trainingsverteilung durch Umgewichtung auf die Testverteilung abgebildet wird. Es werden kombinierte Modelle zum Lernen mit verschiedenen Trainings- und Testverteilungen untersucht, für deren Schätzung nur ein einziges Optimierungsproblem gelöst werden muss. Die kombinierten Modelle können mit zwei Optimierungsschritten approximiert werden und dadurch kann fast jedes gängige Vorhersagemodell so erweitert werden, dass verzerrte Trainingsverteilungen korrigiert werden. In Fallstudien zu Email-Spam-Filterung, HIV-Therapieempfehlung, Zielgruppenmarketing und anderen Anwendungen werden die neuen Modelle mit Referenzmethoden verglichen.
Honkomp, Nora. "Exploring shifts in migration phenology and breeding distribution of declining North American avian aerial insectivores." Kent State University Honors College / OhioLINK, 2021. http://rave.ohiolink.edu/etdc/view?acc_num=ksuhonors1620686253404077.
Повний текст джерелаAnnand, Colin T. "Scaling Relations as Cognitive Dipsticks: Distribution Analysis of Contextually Driven Performance Shifts in Three Linguistic Tasks." University of Cincinnati / OhioLINK, 2017. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1504796379537155.
Повний текст джерелаLättman, Håkan. "Description of spatial and temporal distributions of epiphytic lichens." Licentiate thesis, Linköping University, Linköping University, Department of Physics, Chemistry and Biology, 2008. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-11561.
Повний текст джерелаLichens are, in most cases, sensitive to anthropogenic factors such as air pollution, global warming, forestry and fragmentation. Two studies are included in this thesis. The first is an evaluation of the importance of old oak for the rare epiphytic lichen Cliostomum corrugatum (Ach.) Fr. This study analysed whether C. corrugatum was limited by dispersal or restricted to tree stands with an unbroken continuity or the substrate old oaks. The results provide evidence that the investigated five populations in Östergötland, Sweden, of C. corrugatum exhibit substantial gene flow, an effective dispersal and a small genetic variation between the sites. Most of the genetic variation was within the populations. Thus, C. corrugatum is more dependent of the substrate old oaks, rather than limited by dispersal. The second study investigated possible range shift of some common macrolichens, due to global warming, from 64 sites in southern Sweden comparing the two years 1986 and 2003. The centroid of three lichen species had moved a significant distance, all in a north east direction: Hypogymnia physodes (L.) Nyl. and Vulpicida pinastri (Scop.) J.-E. Mattsson and M. J. Lai on the tree species Juniperus communis L. (50 and 151 km, respectively) and H. physodes on Pinus sylvestris L. (41 km). Considering also the non-significant cases, there is strong evidence for a prevailing NE direction of centroid movement.
Sng, Sin Hie. "Radio channel modeling for mobile ad hoc wireless networks." Thesis, Monterey, Calif. : Springfield, Va. : Naval Postgraduate School ; Available from National Technical Information Service, 2004. http://library.nps.navy.mil/uhtbin/hyperion/04Jun%5FSng.pdf.
Повний текст джерелаThesis advisor(s): Murali Tummala, Roberto Cristi. Includes bibliographical references (p. 71). Also available online.
AL, SALEH-MAHROUSSEH SALOUA. "Calcul relativiste, en electrodynamique quantique, de la diffusion compton sur un electron lie." Clermont-Ferrand 2, 1988. http://www.theses.fr/1988CLF21098.
Повний текст джерелаPit, Vincent. "Codage du flot géodésique sur les surfaces hyperboliques de volume fini." Phd thesis, Université Sciences et Technologies - Bordeaux I, 2010. http://tel.archives-ouvertes.fr/tel-00553138.
Повний текст джерелаLowry, Sonia L. "Analysis of statnamic load test data using a load shed distribution model." [Tampa, Fla.] : University of South Florida, 2005. http://purl.fcla.edu/fcla/etd/SFE0001238.
Повний текст джерелаJönsson, Emil. "Pressure distributions on a valve disk at different opening angles and their impacts on the shaft." Thesis, Karlstads universitet, Fakulteten för hälsa, natur- och teknikvetenskap (from 2013), 2016. http://urn.kb.se/resolve?urn=urn:nbn:se:kau:diva-44367.
Повний текст джерелаBarrachina, Civera Xavier. "Distributional chaos of C0-semigroups of operators." Doctoral thesis, Universitat Politècnica de València, 2013. http://hdl.handle.net/10251/28241.
Повний текст джерелаBarrachina Civera, X. (2013). Distributional chaos of C0-semigroups of operators [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/28241
TESIS
Premiado
Silva, Libania Araújo. "Estrutura produtiva, desigualdade regional e distribuição de renda : dois ensaios para o estado da Bahia." Pós-Graduação em Economia, 2018. http://ri.ufs.br/jspui/handle/riufs/8694.
Повний текст джерелаA presente dissertação é composta por dois ensaios independentes que abordam dois tipos de desigualdade no estado da Bahia que, embora estejam inter-relacionados, possuem caráter distinto. Assim, o primeiro ensaio buscou compreender a dinâmica das disparidades produtiva e de renda entre as microrregiões do estado da Bahia e decompor o crescimento regional do emprego entre os anos 2004 e 2014, utilizando, respectivamente, o coeficiente de Williamson e a análise diferencial-estrutural. Os resultados obtidos revelaram que houve redução da desigualdade inter-regional de renda no território baiano, bem como a realocação do emprego formal gerado pelas atividades produtivas em suas microrregiões. Por sua vez, o segundo ensaio apresenta a discussão acerca da influência da estrutura produtiva sobre as alterações na distribuição interpessoal de renda no estado da Bahia. Nesse caso, a análise está centrada na aplicação do modelo de insumo-produto de Leontief-Miyazawa (1976), a partir do qual foram estimados os impactos de mudanças exógenas sobre a renda. Para tanto, foram utilizados os dados da matriz insumo-produto da Bahia para o ano-base de 2009 (PEROBELLI et al., 2015), bem como as informações de consumo e rendimento das famílias obtidas, respectivamente, por meio da Pesquisa de Orçamento Familiares (POF 2002/2003) e da Pesquisa Nacional por Amostra de Domicílios (PNAD 2009). Na sequência, foram realizadas duas simulações visando analisar o impacto da redução nas transferências de renda (benefícios previdenciários e programas de transferências direta), respectivamente, às famílias pobres e extremamente pobres, e calculados os indicadores de desigualdade de Gini. Os resultados obtidos sugerem que os setores produtivos possuem capacidades distintas entre si de provocar mudanças sobre a distribuição de renda, e que as transferências governamentais atuam como redutores das desigualdades de rendimentos no estado.
São Cristóvão, SE
Shiell, Glenn Raymond. "The spatial distribution and temporal shifts in the biology of Holothuria whitmaei Bell [Echinodermata: Holothuroidea], Ningaloo Reef, Western Australia." University of Western Australia. School of Animal Biology, 2006. http://theses.library.uwa.edu.au/adt-WU2006.0081.
Повний текст джерелаShiell, Glenn Raymond. "The spatial distribution and temporal shifts in the biology of Holothuria whitmaei Bell [Echinodermata: Holothuroidea], Ningaloo Reef, Western Australia /." Connect to this title, 2005. http://theses.library.uwa.edu.au/adt-WU2006.0081.
Повний текст джерелаMarshall, A. M. "Understanding shifts in body size distributions : a comparative study of the impacts of fishing and climate on North Sea demersal fishes." Thesis, University of Sheffield, 2017. http://etheses.whiterose.ac.uk/18955/.
Повний текст джерелаExcoffier, Mathilde. "Chance-Constrained Programming Approaches for Staffing and Shift-Scheduling Problems with Uncertain Forecasts : application to Call Centers." Thesis, Paris 11, 2015. http://www.theses.fr/2015PA112244/document.
Повний текст джерелаThe staffing and shift-scheduling problems in call centers consist in deciding how many agents handling the calls should be assigned to work during a given period in order to reach the required Quality of Service and minimize the costs. These problems are subject to a growing interest, both for their interesting theoritical formulation and their possible applicative effects. This thesis aims at proposing chance-constrained approaches considering uncertainty on demand forecasts.First, this thesis proposes a model solving the problems in one step through a joint chance-constrained stochastic program, providing a cost-reducing solution. A continuous-based approach leading to an easily-tractable optimization program is formulated with random variables following continuous distributions, a new continuous relation between arrival rates and theoritical real agent numbers and constraint linearizations. The global risk level is dynamically shared among the periods during the optimization process, providing reduced-cost solution. The resulting solutions respect the targeted risk level while reducing the cost compared to other approaches.Moreover, this model is extended so that it provides a better representation of real situations. First, the queuing system model is improved and consider the limited patience of customers. Second, another formulation of uncertainty is proposed so that the period correlation is considered.Finally, another uncertainty representation is proposed. The distributionally robust approach provides a formulation while assuming that the correct probability distribution is unknown and belongs to a set of possible distributions defined by given mean and variance. The problem is formulated with a joint chance constraint. The risk at each period is a decision variable to be optimized. A deterministic equivalent problem is proposed. An easily-tractable mixed-integer linear formulation is obtained through piecewise linearizations
Noorani, Mohd Salmi Md. "Spectral and distributional problems for homogeneous extensions of dynamical systems and the rate of mixing of two-dimensional Markov shifts." Thesis, University of Warwick, 1993. http://wrap.warwick.ac.uk/110749/.
Повний текст джерелаBodin, Jeanne. "Observed changes in mountain vegetation of the Alps during the XXth century - Role of climate and land-use changes." Phd thesis, Université Henri Poincaré - Nancy I, 2010. http://tel.archives-ouvertes.fr/tel-00592144.
Повний текст джерелаHussain, Zahir M. "Adaptive instantaneous frequency estimation: Techniques and algorithms." Thesis, Queensland University of Technology, 2002. https://eprints.qut.edu.au/36137/7/36137_Digitised%20Thesis.pdf.
Повний текст джерелаFomicheva, Marina. "The Role of human reference translation in machine translation evaluation." Doctoral thesis, Universitat Pompeu Fabra, 2017. http://hdl.handle.net/10803/404987.
Повний текст джерелаTanto los métodos manuales como los automáticos para la evaluación de la Traducción Automática (TA) dependen en gran medida de la traducción humana profesional. En la evaluación manual, la traducción humana se utiliza a menudo en lugar del texto original para evitar la necesidad de hablantes bilingües, mientras que la mayoría de las técnicas de evaluación automática miden la similitud entre la TA y una traducción humana (comúnmente llamadas traducción candidato y traducción de referencia), asumiendo que cuanto más cerca están, mayor es la calidad de la TA. A pesar del papel fundamental que juega la traducción de referencia en la evaluación de la calidad de la TA, sus características han sido en gran parte ignoradas. Una propiedad inherente de la traducción profesional es la adaptación del texto original a las expectativas del lector. Como consecuencia, la traducción humana puede ser bastante diferente del texto original, lo cual, como se demostrará a lo largo de este trabajo, tiene un fuerte impacto en los resultados de la evaluación de la TA. El primer objetivo de nuestra investigación fue evaluar los efectos del uso de la traducción humana como punto de referencia para la evaluación de la TA. Para lograr este objetivo, comenzamos con una discusión teórica sobre la relación entre textos originales y traducidos. Se identificó la presencia de cambios de traducción opcionales como una de las características fundamentales de la traducción humana. Se analizó el impacto de estos cambios en la evaluación automática y manual de la TA demostrándose en ambos casos que la evaluación está fuertemente sesgada por la referencia proporcionada. El segundo objetivo de nuestro trabajo fue mejorar la precisión de la evaluación automática medida en términos de correlación con los juicios humanos. Dadas las limitaciones de la evaluación basada en la referencia discutidas en la primera parte del trabajo, en lugar de enfocarnos en la similitud, nos concentramos en el impacto de las diferencias entre la TA y la traducción de referencia buscando criterios que permitiesen distinguir entre variación lingüística aceptable y desviaciones inducidas por los errores de TA. En primer lugar, exploramos el uso del contexto sintáctico local para validar las coincidencias entre palabras candidato y de referencia. En segundo lugar, para compensar la falta de información sobre los segmentos de la TA para los cuales no se encontró ninguna relación con la traducción de referencia, introdujimos características orientadas a la fluidez de la TA en la evaluación basada en la referencia. Implementamos nuestro enfoque como una familia de métricas de evaluación automática que mostraron un rendimiento altamente competitivo en una serie de conocidas campañas de evaluación de la TA.
Jakel, Roland. "Das neue Kontaktmodell in Mechanica WF 4.0 mit Reibung : Theoretische Grundlagen und Anwendungsbeispiele." Universitätsbibliothek Chemnitz, 2009. http://nbn-resolving.de/urn:nbn:de:bsz:ch1-200900787.
Повний текст джерелаBodnarova, Adriana. "Texture analysis for automatic visual inspection and flaw detection in textiles." Thesis, Queensland University of Technology, 2000.
Знайти повний текст джерелаFang, Chih-Shan, and 方之珊. "Security of Differential Phase Shift Quantum Key Distribution." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/8hkctt.
Повний текст джерела國立臺灣大學
物理學研究所
106
In this thesis I discuss one of the quantum key distribution (QKD) protocols, the differential phase shift (DPS) protocol, which is designed mainly for the weak coherent state source, and is considered secure under photon number splitting attack. Starting from a brief introduction to the elements needed in the field of QKD, we reviewed the proof of single photon DPS based on Shor-Preskill approach and coherent state DPS based on complementarity approach. Then key rates for DPS QKD with various sources including single photon, parametric down-conversion and coherent state are calculated. Finally, a short conclusion is given and further development is proposed.
Chang, Chin-Hsuan, and 張進宣. "Toward Demonstration of Differential Phase Shift Quantum Key Distribution." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/4mw2ct.
Повний текст джерела