mlsanityData to 7 Oct 2026

Causal Discovery from Conditionally Stationary Time Series

arXiv 2110.06257 ↗First posted 12 Oct 2021cs.LG stat.MLcited by 11

Not on the map. Why?

Abstract (arXiv)

Causal discovery, i.e., inferring underlying causal relationships from observational data, is highly challenging for AI systems. In a time series modeling context, traditional causal discovery methods mainly consider constrained scenarios with fully observed variables and/or data from stationary time-series. We develop a causal discovery approach to handle a wide class of nonstationary time series that are conditionally stationary, where the nonstationary behaviour is modeled as stationarity conditioned on a set of latent state variables. Named State-Dependent Causal Inference (SDCI), our approach is able to recover the underlying causal dependencies, with provable identifiablity for the state-dependent causal structures. Empirical experiments on nonlinear particle interaction data and gene regulatory networks demonstrate SDCI's superior performance over baseline causal discovery methods. Improved results over non-causal RNNs on modeling NBA player movements demonstrate the potential of our method and motivate the use of causality-driven methods for forecasting.

Key facts. Cited by 11 papers in the map (data through 7 Oct 2026). Builds on 30 of its 87 references that are in the map. First posted on arXiv 12 Oct 2021.

Builds on

30 of 87 references.

References · 30

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  4. 2013Auto-Encoding Variational Bayescited by 3,980
  5. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,708
  6. 2016The Concrete Distribution: A Continuous Relaxation of Discrete Random Variablescited by 1,135
  7. 2015Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)cited by 1,061
  8. 2018DAGs with NO TEARS: Continuous Optimization for Structure Learningcited by 745
  9. 2019CLEVRER: CoLlision Events for Video REpresentation and Reasoningcited by 475
  10. 2015A Recurrent Latent Variable Model for Sequential Datacited by 362
  11. 2018Neural Relational Inference for Interacting Systemscited by 339
  12. 2019DAG-GNN: DAG Structure Learning with Graph Neural Networkscited by 329
18 more
  1. 2017A Structured Self-attentive Sentence Embeddingcited by 318
  2. 2016Unsupervised Feature Extraction by Time-Contrastive Learning and Nonlinear ICAcited by 270
  3. 2013Causal Discovery with Continuous Additive Noise Modelscited by 252
  4. 2020DYNOTEARS: Structure Learning from Time-Series Datacited by 198
  5. 2019Causal Discovery from Heterogeneous/Nonstationary Data with Independent Changescited by 166
  6. 2019CATER: A diagnostic dataset for Compositional Actions and TEmporal Reasoningcited by 137
  7. 2021Counterfactual Generative Networkscited by 111
  8. 2020Amortized Causal Discovery: Learning to Infer Causal Graphs from Time-Series Datacited by 99
  9. 2021Learning Temporally Causal Latent Processes from General Temporal Datacited by 96
  10. 2020Causal Discovery in Physical Systems from Videoscited by 77
  11. 2020Hidden Markov Nonlinear ICA: Unsupervised Learning from Nonstationary Time Seriescited by 76
  12. 2019Causal Discovery with General Non-Linear Relationships Using Non-Linear ICAcited by 58
  13. 2018Causal Discovery in the Presence of Missing Datacited by 46
  14. 2019Causal Discovery and Forecasting in Nonstationary Environments with State-Space Modelscited by 45
  15. 2020Reconstructing regime-dependent causal relationships from observational time seriescited by 28
  16. 2018Multiple Causal Inference with Latent Confoundingcited by 14
  17. 2021Neural Relational Inference with Efficient Message Passing Mechanismscited by 10
  18. 2017Causal Discovery in the Presence of Measurement Error: Identifiability Conditionscited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 3; 2024: 1; 2025: 1; 2026: 6.

11 papers.

  1. 2023Case Studies of Causal Discovery from IT Monitoring Time Seriescited by 14
  2. 2023Evaluating Temporal Observation-Based Causal Discovery Techniques Applied to Road Driver Behaviourcited by 5
  3. 2024TS-CausalNN: Learning Temporal Causal Relations from Non-linear Non-stationary Time Series Datacited by 1
  4. 2025Generating Hypotheses of Dynamic Causal Graphs in Neuroscience: Leveraging Generative Factor Models of Observed Time Seriescited by 1
  5. 2026End-to-End Identifiable and Consistent Recurrent Switching Dynamical Systemscited by 1
  6. 2023GDBN: a Graph Neural Network Approach to Dynamic Bayesian Networkcited by 0
  7. 2026Distributed Dynamic Invariant Causal Prediction in Environmental Time Seriescited by 0
  8. 2026TTCD:Transformer Integrated Temporal Causal Discovery from Non-Stationary Time Series Datacited by 0
  9. 2026Hierarchical Causal Abduction: A Foundation Framework for Explainable Model Predictive Controlcited by 0
  10. 2026Identifiable Markov Switching Models with Instantaneous Effects and Exponential Familiescited by 0
  11. 2026Beyond Stationarity in Time Series: Discovering Causal Structures and Latent Regimes via Markov Blanketscited by 0