Disentangling identifiable features from noisy data with structured nonlinear ICA - Archive ouverte HAL Accéder directement au contenu
Proceedings/Recueil Des Communications Année : 2021

Disentangling identifiable features from noisy data with structured nonlinear ICA

Résumé

We introduce a new general identifiable framework for principled disentanglement referred to as Structured Nonlinear Independent Component Analysis (SNICA). Our contribution is to extend the identifiability theory of deep generative models for a very broad class of structured models. While previous works have shown identifiability for specific classes of time-series models, our theorems extend this to more general temporal structures as well as to models with more complex structures such as spatial dependencies. In particular, we establish the major result that identifiability for this framework holds even in the presence of noise of unknown distribution. The SNICA setting therefore subsumes all the existing nonlinear ICA models for time-series and also allows for new much richer identifiable models. Finally, as an example of our framework's flexibility, we introduce the first nonlinear ICA model for time-series that combines the following very useful properties: it accounts for both nonstationarity and autocorrelation in a fully unsupervised setting; performs dimensionality reduction; models hidden states; and enables principled estimation and inference by variational maximum-likelihood.
Fichier principal
Vignette du fichier
Disentangling.pdf (1.56 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03301596 , version 1 (27-07-2021)

Identifiants

  • HAL Id : hal-03301596 , version 1

Citer

Hermanni Hälvä, Sylvain Le Corff, Luc Lehéricy, Jonathan So, Yongjie Zhu, et al.. Disentangling identifiable features from noisy data with structured nonlinear ICA. Advances in Neural Information Processing Systems (NeurIPS), 34, 2021. ⟨hal-03301596⟩
31 Consultations
65 Téléchargements

Partager

Gmail Facebook X LinkedIn More