mlsanityData to 7 Oct 2026

Causal Temporal Representation Learning with Nonstationary Sparse Transition

arXiv 2409.03142 ↗First posted 5 Sep 2024cs.LG stat.MLcited by 12

Not on the map. Why?

Abstract (arXiv)

Causal Temporal Representation Learning (Ctrl) methods aim to identify the temporal causal dynamics of complex nonstationary temporal sequences. Despite the success of existing Ctrl methods, they require either directly observing the domain variables or assuming a Markov prior on them. Such requirements limit the application of these methods in real-world scenarios when we do not have such prior knowledge of the domain variables. To address this problem, this work adopts a sparse transition assumption, aligned with intuitive human understanding, and presents identifiability results from a theoretical perspective. In particular, we explore under what conditions on the significance of the variability of the transitions we can build a model to identify the distribution shifts. Based on the theoretical result, we introduce a novel framework, Causal Temporal Representation Learning with Nonstationary Sparse Transition (CtrlNS), designed to leverage the constraints on transition sparsity and conditional independence to reliably identify both distribution shifts and latent factors. Our experimental evaluations on synthetic and real-world datasets demonstrate significant improvements over existing baselines, highlighting the effectiveness of our approach.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Builds on 45 of its 73 references that are in the map. First posted on arXiv 5 Sep 2024.

Builds on

45 of 73 references.

References · 45

  1. 2018Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representationscited by 929
  2. 2019Variational Autoencoders and Nonlinear ICA: A Unifying Frameworkcited by 494
  3. 2018Nonlinear ICA Using Auxiliary Variables and Generalized Contrastive Learningcited by 278
  4. 2016Unsupervised Feature Extraction by Time-Contrastive Learning and Nonlinear ICAcited by 270
  5. 2019Cross-task weakly supervised learning from instructional videoscited by 199
  6. 2021Disentanglement via Mechanism Sparsity Regularization: A New Principle for Nonlinear ICAcited by 158
  7. 2022CITRIS: Causal Identifiability from Temporal Intervened Sequencescited by 132
  8. 2020Towards Nonlinear Disentanglement in Natural Data with Temporal Sparse Codingcited by 121
  9. 2018Disentangled Sequential Autoencodercited by 117
  10. 2021Learning Temporally Causal Latent Processes from General Temporal Datacited by 96
  11. 2020Disentanglement by Nonlinear ICA with General Incompressible-flow Networks (GIN)cited by 91
  12. 2016Hierarchical Deep Temporal Models for Group Activity Recognitioncited by 89
33 more
  1. 2020Actor-Transformers for Group Activity Recognitioncited by 84
  2. 2022On the Identifiability of Nonlinear ICA: Sparsity and Beyondcited by 84
  3. 2019D3TW: Discriminative Differentiable Dynamic Time Warping for Weakly Supervised Action Alignment and Segmentationcited by 83
  4. 2020Hidden Markov Nonlinear ICA: Unsupervised Learning from Nonstationary Time Seriescited by 76
  5. 2018Weakly-Supervised Action Segmentation with Iterative Soft Boundary Assignmentcited by 74
  6. 2016Connectionist Temporal Modeling for Weakly Supervised Action Labelingcited by 70
  7. 2017Weakly Supervised Action Learning with RNN based Fine-to-coarse Modelingcited by 70
  8. 2018NeuralNetwork-Viterbi: A Framework for Weakly Supervised Video Learningcited by 67
  9. 2024Causal Representation Learning from Multiple Distributions: A General Settingcited by 67
  10. 2021Disentangling Identifiable Features from Noisy Data with Structured Nonlinear ICAcited by 64
  11. 2022Temporally Disentangled Representation Learningcited by 59
  12. 2019Weakly Supervised Energy-Based Learning for Action Segmentationcited by 53
  13. 2022Causal Representation Learning for Instantaneous and Temporal Effects in Interactive Systemscited by 50
  14. 2022Latent Hierarchical Causal Structure Discovery with Rank Constraintscited by 50
  15. 2014Weakly Supervised Action Labeling in Videos Under Ordering Constraintscited by 48
  16. 2016Weakly supervised learning of actions from transcriptscited by 46
  17. 2021AdaRL: What, Where, and How to Adapt in Transfer Reinforcement Learningcited by 46
  18. 2024Nonparametric Partial Disentanglement via Mechanism Sparsity: Sparse Actions, Interventions and Sparse Temporal Dependenciescited by 39
  19. 2019A Hybrid RNN-HMM Approach for Weakly Supervised Temporal Action Segmentationcited by 36
  20. 2024A Sparsity Principle for Partially Observable Causal Representation Learningcited by 34
  21. 2023Generalizing Nonlinear ICA Beyond Structural Sparsitycited by 33
  22. 2019Fast Weakly Supervised Action Segmentation Using Mutual Consistencycited by 30
  23. 2022Factored Adaptation for Non-Stationary Reinforcement Learningcited by 30
  24. 2021Learning Discriminative Prototypes with Dynamic Time Warpingcited by 23
  25. 2021Long Short View Feature Decomposition via Contrastive Video Representation Learningcited by 21
  26. 2023Temporally Disentangled Representation Learning under Unknown Nonstationaritycited by 21
  27. 2022Detector-Free Weakly Supervised Group Activity Recognitioncited by 20
  28. 2024CaRiNG: Learning Temporal Causal Representation under Non-Invertible Generation Processcited by 20
  29. 2021FIFA: Fast Inference Approximation for Action Segmentationcited by 16
  30. 2024Efficient and Effective Weakly-Supervised Action Segmentation via Action-Transition-Aware Boundary Alignmentcited by 13
  31. 2014Marginal likelihood and model selection for Gaussian latent tree and forest modelscited by 12
  32. 2023On the Identifiability of Switching Dynamical Systemscited by 9
  33. 2022Transformers in Action: Weakly Supervised Action Segmentationcited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 1; 2025: 4; 2026: 7.

12 papers.

  1. 2025Thought Communication in Multiagent Collaborationcited by 28
  2. 2025LLM Interpretability with Identifiable Temporal-Instantaneous Representationcited by 4
  3. 2026TRACE: Trajectory Recovery for Continuous Mechanism Evolution in Causal Representation Learningcited by 4
  4. 2024Unsupervised Representation Learning from Sparse Transformation Analysiscited by 2
  5. 2026Causal Representation Meets Stochastic Modeling under Generic Geometrycited by 2
  6. 2026Hierarchical Action Learning for Weakly-Supervised Action Segmentationcited by 1
  7. 2026Identifiability of Potentially Degenerate Gaussian Mixture Models With Piecewise Affine Mixingcited by 1
  8. 2026MOSAIC: Module Discovery via Sparse Additive Identifiable Causal Learning for Scientific Time Seriescited by 1
  9. 2026End-to-End Identifiable and Consistent Recurrent Switching Dynamical Systemscited by 1
  10. 2025Infant Cry Detection Using Causal Temporal Representationcited by 0
  11. 2025Linear Causal Representation Learning by Topological Ordering, Pruning, and Disentanglementcited by 0
  12. 2026Disentangling Continuous-Time Latent Dynamics: Identifiability of Latent SDEs via Diffusion Shiftscited by 0