Disentangled Representation Learning

Overview

Disentangled representation learning studies how to encode separate factors—such as content and style, identity and viewpoint, or disease and acquisition conditions—in distinct parts of a representation. Common approaches include variational autoencoders, contrastive objectives, sparse or orthogonal subspaces, causal models, and symmetry-based methods; many applications use the resulting factors for editing, robust prediction, or transfer across domains. Work now spans images, speech, graphs, time series, medicine, and robotics, with increasing attention to interventions and practical controllability. A central tension remains: factors are generally not identifiable from unlabelled data alone without assumptions or useful supervision, and standard disentanglement metrics can fail when their assumptions do not hold.

Latest papers

CardsList

245 papers10 papers in the last four weeks, down 63% on the four weeks before. 0.1% of all new papers.

  1. Post-hoc Disentanglement of Textual and Acoustic Features in Pre-trained Speech Encoders

    Oct 3, 2024Hosein Mohebbi, Grzegorz Chrupała, Willem Zuidema +2Representation DisentanglementDisentangled Representation Learning

  2. Beyond Spectral Decomposition: Bayesian Contrastive Learning and its Non-negative Formulation via Factor Analysis

    Jul 31, 2024Zhibin Duan, Tiansheng Wen, Yifei Wang +3Contrastive LearningDisentangled Representation Learning

  3. InfoDiffusion: Representation Learning Using Information Maximizing Diffusion Models

    Jun 14, 2023Yingheng Wang, Yair Schiff, Aaron Gokaslan +4Diffusion ModelsDisentangled Representation Learning

  4. Disentanglement with Biological Constraints: A Theory of Functional Cell Types

    Sep 30, 2022James C. R. Whittington, Will Dorrell, Surya Ganguli +1Disentangled Representation LearningNeural Representation Geometry