Aapo Hyvärinen (Inria, University of Helsinki) est l’animateur du séminaire DATAIA du 9 octobre qui se concentrera sur un des plus grands défis de l'apprentissage machine : l'apprentissage non supervisé.

Il illustrera et discutera de l'analyse non linéaire en composantes indépendantes (ACI), la méthode qui permet de résoudre ce problème; de ses contributions au sujet; et du lien entre l'ACI non linéaire et les encodeurs à variation automatique (VAE).

"Unsupervised learning, in particular learning general nonlinear representations, is one of the deepest problems in machine learning. Estimating latent quantities in a generative model provides a principled framework, and has been successfully used in the linear case, e.g. with independent component analysis (ICA) and sparse coding. However, extending ICA to the nonlinear case has proven to be extremely difficult: A straight-forward extension is unidentifiable, i.e. it is not possible to recover those latent components that actually generated the data. Here, we show that this problem can be solved by using additional information either in the form of temporal structure or an additional, auxiliary variable. As a first approach, we formulate self-supervised learning schemes which are similar to those heuristically proposed in computer vision. Our main contribution is to provide a rigorous theoretical framework for such self-supervised algorithms, proving that they are able to solve the nonlinear ICA problem. We further show how a connection between nonlinear ICA and variational autoencoders (VAE): While ordinary VAE suffers from the lack of identifiability, conditioning by auxiliary variables leads to identifiability and provides another method for learning nonlinear ICA."

Le séminaire aura lieu le 9 octobre 2019 de 14h00 à 16h00 au centre Inria Saclay.

Inscriptions gratuites mais obligatoires dans la limite des places disponibles
Pour des raisons de sécurité, toute personne non-inscrite ne pourra accéder au lieu du séminaire

Autres séminaires DATAIA