Disentanglement Analysis in Deep Latent Variable Models Matching Aggregate Posterior Distributions
Deep latent variable models (DLVMs) are designed to learn meaningful representations in an unsupervised manner, such that the hidden explanatory factors are interpretable by independent latent variables (aka disentanglement). The variational autoencoder (VAE) is a popular DLVM widely studied in disentanglement analysis due to the modeling of the posterior distribution using a factorized Gaussian distribution that encourages the alignment of the latent factors with the latent axes. Several metrics have been proposed recently, assuming that the latent variables explaining the variation in data are aligned with the latent axes (cardinal directions). However, there are other DLVMs, such as the AAE and WAE-MMD (matching the aggregate posterior to the prior), where the latent variables might not be aligned with the latent axes. In this work, we propose a statistical method to evaluate disentanglement for any DLVMs in general. The proposed technique discovers the latent vectors representing the generative factors of a dataset that can be different from the cardinal latent axes. We empirically demonstrate the advantage of the method on two datasets.
Code (0)
등록된 구현이 없습니다.
Tasks
DisentanglementSimilar Papers 제목 키워드 기반
Structured Coupling for Flow Matching
Standard flow matching scales well but typically relies on an unstructured source distribution, limiting its ability to learn interpretable latent structure. Latent-variable models, by contrast, capture structure but oft…
Representation LearningLinear causal disentanglement via higher-order cumulants
Linear causal disentanglement is a recent method in causal representation learning to describe a collection of observed variables via latent variables with causal dependencies between them. It can be viewed as a generali…
DisentanglementRepresentation LearningTensor DecompositionDisentanglement in Difference: Directly Learning Semantically Disentangled Representations by Maximizing Inter-Factor Differences
In this study, Disentanglement in Difference(DiD) is proposed to address the inherent inconsistency between the statistical independence of latent variables and the goal of semantic disentanglement in disentanglement rep…
DisentanglementRepresentation LearningLinear Causal Disentanglement via Interventions
Causal disentanglement seeks a representation of data involving latent variables that relate to one another via a causal model. A representation is identifiable if both the latent model and the transformation from latent…
DisentanglementDisentanglement via Mechanism Sparsity Regularization: A New Principle for Nonlinear ICA
This work introduces a novel principle we call disentanglement via mechanism sparsity regularization, which can be applied when the latent factors of interest depend sparsely on past latent factors and/or observed auxili…
DisentanglementRepresentation Learning