Добірка наукової літератури з теми "Apprentissage automatique guidé par la physique"
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся зі списками актуальних статей, книг, дисертацій, тез та інших наукових джерел на тему "Apprentissage automatique guidé par la physique".
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.
Дисертації з теми "Apprentissage automatique guidé par la physique":
Brandão, Eduardo. "Complexity Methods in Physics-Guided Machine Learning." Electronic Thesis or Diss., Saint-Etienne, 2023. http://www.theses.fr/2023STET0062.
Complexity is easy to recognize but difficult to define: there are a host of measures of complexity, each relevant for a particular application.In Surface engineering, self-organization drives the formation of patterns on matter by femtosecond laser irradiation, which have important biomedical applications. Pattern formation details are not fully understood. In work leading to two publications [1,2], via a complexity argument and a physics-guided machine learning framework, we show that the severely constrained problem of learning the laser-matter interaction with few data and partial physical knowledge is well-posed in this context. Our model allows us to make useful predictions and suggests physical insights.In another contribution [3] we propose a new formulation of the Minimum Description Length principle, defining model and data complexity in a single step, by taking into account signal and noise in training data. Experiments indicate that Neural Network classifiers that generalize well follow this principle.In unpublished work, we propose Taylor entropy, a novel measure of dynamical system complexity which can be estimated via a single SEM image. This approach could facilitate learning the physical process in new materials through domain adaptation.This thesis paves the way for a unified representation of complexity in data and physical knowledge, which can be used in the context of Physics-guided machine learning.[1] Brandao, Eduardo, et al. "Learning PDE to model self-organization of matter." Entropy 24.8 (2022): 1096.[2] Brandao, Eduardo, et al. "Learning Complexity to Guide Light-Induced Self-Organized Nanopatterns." Physical Review Letters 130.22 (2023): 226201.[3] Brandao, Eduardo, et al. "Is My Neural Net Driven by the MDL Principle?." Joint European Conference on Machine Learning and Knowledge Discovery in Databases. Cham: Springer Nature Switzerland, 2023
Jiao, Yunlong. "Pronostic moléculaire basé sur l'ordre des gènes et découverte de biomarqueurs guidé par des réseaux pour le cancer du sein." Thesis, Paris Sciences et Lettres (ComUE), 2017. http://www.theses.fr/2017PSLEM027/document.
Breast cancer is the second most common cancer worldwide and the leading cause of women's death from cancer. Improving cancer prognosis has been one of the problems of primary interest towards better clinical management and treatment decision making for cancer patients. With the rapid advancement of genomic profiling technologies in the past decades, easy availability of a substantial amount of genomic data for medical research has been motivating the currently popular trend of using computational tools, especially machine learning in the era of data science, to discover molecular biomarkers regarding prognosis improvement. This thesis is conceived following two lines of approaches intended to address two major challenges arising in genomic data analysis for breast cancer prognosis from a methodological standpoint of machine learning: rank-based approaches for improved molecular prognosis and network-guided approaches for enhanced biomarker discovery. Furthermore, the methodologies developed and investigated in this thesis, pertaining respectively to learning with rank data and learning on graphs, have a significant contribution to several branches of machine learning, concerning applications across but not limited to cancer biology and social choice theory
Houssein, Aya. "Pléthysmographie respiratoire par magnétométrie ˸ Evaluation de la ventilation et de la dépense énergétique à partir d'algorithmes d'apprentissage automatique." Thesis, Rennes, École normale supérieure, 2021. http://www.theses.fr/2021ENSR0026.
Regular physical activity (PA) is essential to maintain and improve health. The quantification of PA has become a major focus in scientific research studying the relationship between PA and its effects on health. PA is generally quantified in terms of energy expenditure (EE). Reference methods used to measure EE are cumbersome and invasive. To overcome the problems associated with the use of reference methods, portable and non-invasive devices have been developed. Among these devices, respiratory magnetometer plethysmography (RMP) has recently developed. PRM is based on the measurement of the longitudinal and transversal thoracic and abdominal distances.The objective of this thesis is to evaluate the ability of PRM to estimate V˙E and EE during low to high intensity PA using machine learning algorithms. The main results of our work demonstrate 1) That RMP is suitable to estimate ˙VE and DE during low to high PA. 2) A nonlinear model is more relevant than a linear model to estimate V˙E. 3) The individualization of the models provides better performance for V˙E and EE estimation.4) RMP can accurately estimate EE at any intensity, including the highest ones. 4) An activity-specific approach is more relevant to estimate EE ,and a step of PA recognition is necessary before EE estimation.Further studies are still needed to evaluate RMP on a large population and under free-living conditions
Buhot, Arnaud. "Etude de propriétés d'apprentissage supervisé et non supervisé par des méthodes de Physique Statistique." Phd thesis, Université Joseph Fourier (Grenoble), 1999. http://tel.archives-ouvertes.fr/tel-00001642.
Garnotel, Maël. "Apport de la reconnaissance des postures et des activités par accélérométrie à la caractérisation du comportement de mouvement chez l’humain : application à l’étude de la transition épidémiologique chez les Peuls." Thesis, Lyon, 2019. https://n2t.net/ark:/47881/m6kk9b3r.
Facing the rise of non-communicable diseases, physical activity and sedentary behavior are a major health issue. Evaluating the synergies of movement behavior dimensions in order to establish its link with health emerges is a key challenge. The development of accelerometry has revolutionized the understanding of these links, traditionally studied using declarative data, associated with well-established biases. The classical accelerometry approach allows continuous measurement over long periods in free living conditions but encounters limitations inherent in signal processing and in the non-linear relationship between accelerometry and energy expenditure to characterize human movement in a satisfactory way. My first objective was to clarify the limits of the current approach and to contribute to the improvement of the phenotyping of movement behavior through new analytic methods relying on the automatic activity recognition of postures and activities. In the second part of my thesis, I applied these new approaches to the study of the Fulani of Senegal, a population in epidemiological transition. My work has clarified the limitations of traditional approaches to accelerometry and the value of activities recognition through automatic learning algorithms to overcome the difficulties encountered. For the first time, they show the contribution of this approach to the detailed characterization of a population's physical activity and sedentary behaviors, in relation to its environment. It should contribute in a useful way to the development of future recommendations that are more appropriate for the general population
Melnyk, Artem. "Perfectionnement des algorithmes de contrôle-commande des robots manipulateur électriques en interaction physique avec leur environnement par une approche bio-inspirée." Thesis, Cergy-Pontoise, 2014. http://www.theses.fr/2014CERG0745/document.
Automated production lines integrate robots which are isolated from workers, so there is no physical interaction between a human and robot. In the near future, a humanoid robot will become a part of the human environment as a companion to help or work with humans. The aspects of coexistence always presuppose physical and social interaction between a robot and a human. In humanoid robotics, further progress depends on knowledge of cognitive mechanisms of interpersonal interaction as robots physically and socially interact with humans. An illustrative example of interpersonal interaction is an act of a handshake that plays a substantial social role. The particularity of this form of interpersonal interaction is that it is based on physical and social couplings which lead to synchronization of motion and efforts. Studying a handshake for robots is interesting as it can expand their behavioral properties for interaction with a human being in more natural way. The first chapter of this thesis presents the state of the art in the fields of social sciences, medicine and humanoid robotics that study the phenomenon of a handshake. The second chapter is dedicated to the physical nature of the phenomenon between humans via quantitative measurements. A new wearable system to measure a handshake was built in Donetsk National Technical University (Ukraine). It consists of a set of several sensors attached to the glove for recording angular velocities and gravitational acceleration of the hand and forces in certain points of hand contact during interaction. The measurement campaigns have shown that there is a phenomenon of mutual synchrony that is preceded by the phase of physical contact which initiates this synchrony. Considering the rhythmic nature of this phenomenon, the controller based on the models of rhythmic neuron of Rowat-Selverston, with learning the frequency during interaction was proposed and studied in the third chapter. Chapter four deals with the experiences of physical human-robot interaction. The experimentations with robot arm Katana show that it is possible for a robot to learn to synchronize its rhythm with rhythms imposed by a human during handshake with the proposed model of a bio-inspired controller. A general conclusion and perspectives summarize and finish this work
Desbordes, Paul. "Méthode de sélection de caractéristiques pronostiques et prédictives basée sur les forêts aléatoires pour le suivi thérapeutique des lésions tumorales par imagerie fonctionnelle TEP." Thesis, Normandie, 2017. http://www.theses.fr/2017NORMR030/document.
Radiomics proposes to combine image features with those extracted from other modalities (clinical, genomic, proteomic) to set up a personalized medicine in the management of cancer. From an initial exam, the objective is to anticipate the survival rate of the patient or the treatment response probability. In medicine, classical statistical methods are generally used, such as theMann-Whitney analysis for predictive studies and analysis of Kaplan-Meier survival curves for prognostic studies. Thus, the increasing number of studied features limits the use of these statistics. We have focused our works on machine learning algorithms and features selection methods. These methods are resistant to large dimensions as well as non-linear relations between features. We proposed two features selection strategy based on random forests. Our methods allowed the selection of subsets of predictive and prognostic features on 2 databases (oesophagus and lung cancers). Our algorithms showed the best classification performances compared to classical statistical methods and other features selection strategies studied
Philippeau, Jérémy. "Apprentissage de similarités pour l'aide à l'organisation de contenus audiovisuels." Toulouse 3, 2009. http://thesesups.ups-tlse.fr/564/.
In the perspective of new usages in the field of the access to audiovisual archives, we have created a semi-automatic system that helps a user to organize audiovisual contents while performing tasks of classification, characterization, identification and ranking. To do so, we propose to use a new vocabulary, different from the one already available in INA documentary notices, to answer needs which can not be easily defined with words. We have conceived a graphical interface based on graph formalism designed to express an organisational task. The digital similarity is a good tool in respect with the handled elements which are informational objects shown on the computer screen and the automatically extracted audio and video low-level features. We have made the choice to estimate the similarity between those elements with a predictive process through a statistical model. Among the numerous existing models, the statistical prediction based on the univaried regression and on support vectors has been chosen. H)
Zou, Long. "Simulation of laser energy deposition with structured light beams in air and machine learning data treatment for LIBS analysis of remote targets." Electronic Thesis or Diss., Institut polytechnique de Paris, 2023. http://www.theses.fr/2023IPPAX053.
The propagation of ultrashort and ultra-intense laser pulses in the atmosphere is increasingly studied.Precise control of the focusing of the laser field and the distribution of light filaments extending beyond the focus is required for many applications, such as the Mars exploration mission on the analysis of chemical elements using Laser Induced Breakdown Spectroscopy (LIBS), the analysis of the composition of air by Light Detection and Ranging (LIDAR) techniques, the triggering and guiding of electric discharges between clouds, or the generation of white light laser by filamentation. A quantitative control of the light pulse properties is very difficult due to the complex nonlinear interaction between the intense laser pulse and the medium. At present, commonly used methods rely on the parameter control of the initial laser output and the feedback of the field at the target position. The high-dimension of the parameter space and the high sensitivity of the results to the initial conditions make the adjustment of the laser field outside the laboratory difficult and inefficient to meet the requirements of practical applications.In this context, this thesis proposes an answer to some of the challenges of long-range femtosecond laser pulse propagation, based on laser pulse modulation scenarios that guarantee to reach an on-target laser field with the desired properties. These scenarios were obtained by combining reverse engineering methods and numerical simulations. We show that different target fields can be easily and efficiently achieved by modulating the laser output field. Whenever possible, the modulation is obtained by simulating the reverse propagation of the target field towards the laser.This thesis focuses on two different objectives of laser field control: the long-range projection in the air of (1) a filament of predefined length, and (2) high intensity.(1) To achieve the first objective, one of the innovations of this thesis consists in introducing a controllable intermediate Bessel-Gauss beam close to the target, and in using a numerical algorithm to propagate this electric field forward in order to obtain the distribution of the filaments at the target point as well as back-propagate the intermediate field to obtain the desired laser output. The obtained laser output parameters are then related to filament features (starting point, length, density), providing a map for the key parameters defining the modulated laser pulse that can be projected onto the desired target field and filament.(2) For the objective of transmitting high intensities at kilometric distances, we examine the nonlinear propagation of circular Airy beams and show that a laser power of a few tens of GW is sufficient to ionize the air and form a short filament at a distance of 1 km, which could facilitate laser operating conditions compared to TW-class lasers used in conventional solutions to project high intensities at these distances.In a separate study, we propose an improvement of the elemental analysis algorithm of LIBS spectrum. The algorithm is applied to the extit{in situ} online detection of KCl and H[dollar]_2[dollar]O content in potash fertilizer by LIBS, in which correlation regression modeling of LIBS spectra is combined with a machine learning algorithm that efficiently extracts the information related to elemental content changes from the complex online collected spectra, which greatly improves the detection speed while ensuring the detection accuracy and further enhances the competitiveness
Yang, Tong. "Constitution et exploitation d’une base de données pour l’enseignement/apprentissage des phrasèmes NAdj du domaine culinaire français auprès d’apprenants non-natifs." Thesis, Paris 3, 2019. http://www.theses.fr/2019PA030049.
This thesis project aims to study the teaching method of FOS (French on Specific Objectives) catering to foreign cooks who come to work in French restaurants or who have chosen catering as a specialty. The objective of our research is therefore to teach the culinary NAdj phrasemas to foreign A2 level learners. The teaching/learning of phraseology is required in specialty languages and the high frequency of NAdj phrasems has caught our attention. Several questions are then addressed: where to find this specific lexicon? How to extract them? By which approach do we teach the selected phrasems? To answer these questions, we made our own corpus Cuisitext - written and oral - and then used NooJ to extract the NAdj phrasems from the corpus. Finally, we have proposed the three approaches to the use of corpora for the teaching/learning of NAdj phrasems: guided inductive approach, deductive approach, pure inductive approach