A Revisit of Total Correlation in Disentangled Variational Auto-Encoder with Partial Disentanglement

ICLR 2026 Conference Submission14393 Authors

Keywords: disentangled variational auto-encoder, partial disentanglement, group-wise independence

TL;DR: We propose the Partially Disentangled VAE, which relaxes full independence into group-wise independence via a partial correlation term, enabling more flexible disentanglement and yielding valuable insights on both synthetic and real-world datasets.

Abstract: A fully disentangled variational auto-encoder (VAE) aims to identify disentangled latent components from observations. However, enforcing full independence between all latent components may be too strict for certain datasets. In some cases, multiple factors may be entangled together in a non-separable manner, or a single independent semantic meaning could be represented by multiple latent components within a higher-dimensional manifold. To address such scenarios with greater flexibility, we develop the Partially Disentangled VAE (PDisVAE), which generalizes the total correlation (TC) term in fully disentangled VAEs to a partial correlation (PC) term. This framework can handle group-wise independence and can naturally reduce to either the standard VAE or the fully disentangled VAE. Validation through three synthetic experiments demonstrates the correctness and practicality of PDisVAE. When applied to real-world datasets, PDisVAE discovers valuable information that is difficult to uncover with fully disentangled VAEs, implying its versatility and effectiveness.

Primary Area: unsupervised, self-supervised, semi-supervised, and supervised representation learning

Submission Number: 14393