Adaptation de domaine

Résumé

L'adaptation de domaine (domain adaptation en anglais) est un champ de recherche de l'apprentissage automatique, et plus précisément de l'apprentissage par transfert. L'objectif est d'effectuer une tâche d'adaptation d'un système d'apprentissage d'un domaine (une distribution de probabilité) source vers un domaine (une distribution de probabilité) cible. Lorsque plusieurs domaines sources sont disponibles, on parle alors d’adaptation de domaine multi-sources. Un peu de formalisme Soit X l'espace d'entrée ou de description des données et Y l'espace de sortie ou l'ensemble d'étiquetage. L'objectif en apprentissage automatique est d'apprendre un modèle mathématique (une hypothèse) h:X\to Y permettant d'affecter une étiquette à une donnée. Ce modèle est appris à partir d'un échantillon d'apprentissage S={(x_i,y_i)}_{i=1}^m \in (X\times Y)^m. En apprentissage supervisé classique (sans adaptation de domaine), on suppose qu

Source officielle

https://fr.wikipedia.org/wiki/Adaptation_de_domaine

À propos de ce résultat

Cette page est générée automatiquement et peut contenir des informations qui ne sont pas correctes, complètes, à jour ou pertinentes par rapport à votre recherche. Il en va de même pour toutes les autres pages de ce site. Veillez à vérifier les informations auprès des sources officielles de l'EPFL.

Publications associées (28)

Publications associées

Chargement

Personnes associées (6)

Personnes associées

Chargement

Unités associées (3)

Unités associées

Chargement

Concepts associés

Chargement

Cours associés (1)

Cours associés

Chargement

Séances de cours associées (1)

Séances de cours associées

Chargement

Domain adaptation using manifold alignment

Maxime Trolliet

Domain adaptation is a major challenge for future remote sensing applications. Both financial and temporal constraints of data acquisition lead to the developing of new techniques able to use knowledge from alternative sources. Different approaches have been developed by considering the statistical properties of images or by modifying already existing classifiers. We propose a intermediary approach of these two kinds of methods by using a manifold alignment technique constrained by similarity between two images. The two images are mapped in a high dimensional latent space which maximizes the proximity of similar elements, thus allowing classification of the images suffering from label scarcity by using the knowledge of the other image. Such a classification offers improvement compared to various used processes.

2012

Chargement

Understanding behavior from neural activity is a fundamental goal in neuroscience. It has practical applications in building robust brain-machine interfaces, human-computer interaction, and assisting patients with neurological disabilities. Despite the ever-growing demand driven by the applications, neural action decoding methods still require laborious manual annotations. Furthermore, a different model must be trained for each new individual, requiring even more annotation and overwhelming the resources of most applications. This thesis proposes a self-supervised way to tackle the unsupervised domain adaptation for the neural action decoding problem. Our method does not require manual annotations and can generalize across individuals. To do that, we propose to use animal motion capture to guide the encoding of neural action representations together with a set of neural and behavioral augmentations. We, therefore, divide the thesis into two parts. For the first part, we propose multiple novel ways to learn robust markerless motion capture systems for animals. In the second part, we then use motion capture to learn an unsupervised neural action decoding algorithm.To start, we present a novel unsupervised markerless 2D pose estimation system. Extracting pose information from laboratory animals requires significant annotation labor. Our model requires only unrealistic renderings of the target animal, without any manual pose labeling. While deep learning models cannot be trained on unrealistic renderings of laboratory animals due to the domain gap, we propose a deformation-based generative network transforming the renderings into realistic images while at the same time keeping the pose information intact. Our model enables us to train a pose estimation network on generated but realistic images without manual annotations.Second, we propose a markerless self-calibrating motion capture system on tethered animals. Although extracting a 2D pose is enough for some applications, 2D pose information is incomplete due to information loss from the camera projection. On the other hand, triangulating 3D poses from multiple images requires camera calibration. This tedious process is not ideal for many laboratory setups, for example, a small model organism \textit{Drosophila}. We model a self-calibrating system that uses the animal as the calibration pattern. To make the system even more reliable, we use the shape and skeleton of the animal as the 3D prior, therefore fixing the inference mistakes of the deep learning based methods.Third, we propose a monocular 3D motion capture algorithm for laboratory animals. Although 3D pose information provides all the necessary information about the movement, multi-camera systems require expensive hardware and multi-view synchronization. We propose a monocular system that can still recover 3D pose without requiring multiple views. We test our model on numerous lab animals, including flies, mice, rats, and macaques. We show that our monocular motion capture system performs reliably and on-par with multi-camera motion capture systems.Lastly, we propose a novel unsupervised domain adaptation method for neural decoding of actions, which uses the previously mentioned animal markerless motion capture systems for domain adaptation. We show that our neural action decoding algorithm can successfully generalize across individuals. Furthermore, our method can provide better neural acti

EPFL2022

Chargement

Learning and leveraging shared domain semantics to counteract visual domain shifts

Róger Bermúdez Chacón

One of the main limitations of artificial intelligence today is its inability to adapt to unforeseen circumstances. Machine Learning (ML), due to its data-driven nature, is particularly susceptible to this. ML relies on observations in order to learn implicit rules about the inputs, outcomes, and relationships among them, so as to solve a task. An unfortunate consequence of learning based on observations, however, is that ML algorithms learn by observing a partial, inevitably skewed version of the world. As a result, ML methods experience difficulties deciding how to properly use their learned experience when the world as they know it changes. Domain adaptation, the paradigm followed in this thesis, is an area of ML research that tackles the above problem. In the domain adaptation setup, a ML problem must be adapted when domain shiftsâor changes in the nature of the dataâoccur. This thesis tackles the domain adaptation problem in the particular context of visual applications, with the ultimate goal of reducing as much as possible the need for human intervention in training ML methods for visual tasks. We study the domain adaptation problem from two different fronts. A first idea is to harness the existing structure of the images. Despite visual differences, structural information related to the semantic content of the image is often preserved in images from different origins. We present a method to leverage those visual correspondencesâeven when the match is imperfectâbased on Multiple Instance Learning, so as to adjust parameters of a ML model to new domains. We also introduce a Self-Supervised ML method that aggregates visual correspondences into a consensus heatmap. As such heatmaps are good unsupervised proxies for real annotations, we use them as supervisory signal. In addition, we propose a Two-Stream U-Net architecture that processes different domains simultaneously. The Two-Stream U-Net combines parameter regularization, distribution matching, and the self-supervised signal from the consensus heatmaps, to bridge the performance gap between models operating on different image domains. The second line of reasoning looks beyond the raw image information and instead maps images to compact latent representations that preserve image semantics. For this, we introduce multiflow networks: a neural Network Architecture Search paradigm that assigns different network capacity to different image domains to extract domain-agnostic latent representations. In the multiflow formalism, domain-specific learnable gates modulate the contribution of different operations to the encoding. The end results are latent encodings from different domains that do not suffer from the domain shift. Our results in biomedical image segmentation, object classification, and object detection, validate the wide range of applicability of the methods introduced in this thesis.

EPFL2020

Chargement

EPFL2022

Learning and leveraging shared domain semantics to counteract visual domain shifts

Róger Bermúdez Chacón

EPFL2020