Representation Identifiability

Overview

Representation identifiability asks whether learned codes correspond to recoverable features of the data or world, rather than one of many equally valid internal coordinate systems. Researchers use nonlinear independent component analysis, multi-view and temporal data, sparsity or orthogonality constraints, and symmetry-breaking interventions to establish when recovery is possible. Guarantees may identify individual factors only up to permutation and scaling, or recover just shared subspaces and invariant geometry, in settings from generative models to neural networks. A central tension is that results depend on assumptions about the data, model, and measurements: identical predictions can arise from different representations, and evaluation metrics can themselves change under function-preserving reparameterizations.

Latest papers

CardsList

44 papers2 papers in the last four weeks, down 85% on the four weeks before. 0.0% of all new papers.

  1. Identifiable but Hard to Estimate: Rotation in Equally Spaced Overcomplete ICA

    Nov 9, 2026Wenjun Xiao, Yinghao Fang, Hanxin Chen +3Independent Component AnalysisSample Complexity

  2. Complete or Sparse: A Tale of Two Identifiabilities

    Oct 14, 2026Junze Zhou, David KlindtSparse CodingRepresentation Identifiability

  3. Identication and Estimation of Nonsymmetric Determinantal Point Processes

    Oct 10, 2026Christian Gouriéroux, Yang LuDeterminantal Point ProcessRepresentation Identifiability

  4. Intervention anchors and scientific verification in synthetic vascular predictive representations

    Oct 8, 2026Lingsen You, Yujun Guo, Xinyu Zhong +4Representation IdentifiabilityPredictive Representation Learning

  5. DSReg: Provably Recovering Individual World Latents without Reconstruction

    Oct 7, 2026Yujia Zheng, David Klindt, Randall Balestriero +1Representation IdentifiabilityLatent World Models

  6. Parameter Symmetries Determine Representational Geometry in Overparameterized Nonlinear Networks

    Sep 30, 2026Marvin Theiss, Lukas Braun, Andrew M. Saxe +1Overparameterized Neural NetworkNeural Representation Geometry

  7. Predictive Self-Supervised Learning Provably Identifies Stochastic Signals under Nuisance

    Sep 29, 2026Fabian A. Mikulasch, Friedemann ZenkeLatent Variable ModelsRepresentation Identifiability

  8. Identifying ODEs from Unstructured Data with Causal Representation Learning

    Sep 29, 2026Alessandro Trenta, Riccardo Massidda, Davide Bacciu +1Causal Representation LearningRepresentation Identifiability

  9. Towards Identifiable Representations under Misspecified Structure

    Sep 27, 2026Yuke Li, Yujia Zheng, Ziyi Chen +2Latent Variable ModelsRepresentation Identifiability

  10. Discovering Physical Representation Languages

    Sep 20, 2026Linzhe Zhang, Changming XuRepresentation Identifiability

  11. Kinks vs. Smoothness: Identifiability of Real Analytic nICA for Laplace-like Sources

    Sep 18, 2026Isaac Manring, Kejun HuangIndependent Component AnalysisRepresentation Identifiability

  12. Converse and Collision-Based Achievability for Node Localization with Hybrid Distance-Spectral Graph Positional Encodings

    Aug 31, 2026Zimo Yan, Yifan Li, Hao Li +4Representation IdentifiabilityPositional Encoding

  13. More Data Cannot Break a Symmetry: Identifiability by Design

    Aug 27, 2026Jing Xu, Christopher KananRepresentation IdentifiabilityParameter Identifiability

  14. Representation Measurements Under Function-Preserving Reparameterizations

    Aug 27, 2026Abdullah KarasanIntrinsic DimensionalityRepresentation Identifiability

  15. Low-Interaction-Rank Learning: Unifying Multiplicative Dual-Encoder Heads

    Aug 12, 2026Zijian Zhao, Sen LiRepresentation IdentifiabilityMultimodal Contrastive Learning

  16. Teacher Supervision over Representation Equivalence Classes

    Jul 3, 2026Sang Il HanRepresentation IdentifiabilityLanguage Model Distillation

  17. Disentangling Continuous-Time Latent Dynamics: Identifiability of Latent SDEs via Diffusion Shifts

    Jun 26, 2026Yuanyuan Wang, Wenjie Wang, Haoxuan Li +2Stochastic Differential EquationsCausal Representation Learning

  18. Unsupervised Disentanglement Without Compromises : How Functional Orthogonality Enforces Identifiability

    Jun 19, 2026Mathieu Cyrille Simon, Pascal Frossard, Christophe De VleeschouwerUnsupervised LearningDisentangled Representation Learning

  19. A Fiber Criterion for Representation Identifiability in Supervised Learning

    May 31, 2026Vasileios SevetlidisRepresentation IdentifiabilityParameter Identifiability