mlsanityData to 7 Oct 2026

Causal Contrastive Learning for Counterfactual Regression Over Time

arXiv 2406.00535 ↗First posted 1 Jun 2024cs.LG stat.MEcited by 7

Not on the map. Why?

Abstract (arXiv)

Estimating treatment effects over time holds significance in various domains, including precision medicine, epidemiology, economy, and marketing. This paper introduces a unique approach to counterfactual regression over time, emphasizing long-term predictions. Distinguishing itself from existing models like Causal Transformer, our approach highlights the efficacy of employing RNNs for long-term forecasting, complemented by Contrastive Predictive Coding (CPC) and Information Maximization (InfoMax). Emphasizing efficiency, we avoid the need for computationally expensive transformers. Leveraging CPC, our method captures long-term dependencies in the presence of time-varying confounders. Notably, recent models have disregarded the importance of invertible representation, compromising identification assumptions. To remedy this, we employ the InfoMax principle, maximizing a lower bound of mutual information between sequence data and its representation. Our method achieves state-of-the-art counterfactual estimation results using both synthetic and real-world data, marking the pioneering incorporation of Contrastive Predictive Encoding in causal inference.

Key facts. Cited by 7 papers in the map (data through 7 Oct 2026). Builds on 52 of its 84 references that are in the map. First posted on arXiv 1 Jun 2024.

Builds on

52 of 84 references.

References · 52

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  2. 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
  3. 2014Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translationcited by 4,016
  4. 2014Unsupervised Domain Adaptation by Backpropagationcited by 1,906
  5. 2019Contrastive Multiview Codingcited by 1,297
  6. 2018Learning deep representations by mutual information estimation and maximizationcited by 1,265
  7. 2019Learning Representations by Maximizing Mutual Information Across Viewscited by 733
  8. 2018MINE: Mutual Information Neural Estimationcited by 680
  9. 2019Data-Efficient Image Recognition with Contrastive Predictive Codingcited by 672
  10. 2017Self-Normalizing Neural Networkscited by 591
  11. 2019On Variational Bounds of Mutual Informationcited by 550
  12. 2019A Theoretical Analysis of Contrastive Unsupervised Representation Learningcited by 530
38 more
  1. 2016f-GAN: Training Generative Neural Samplers using Variational Divergence Minimizationcited by 504
  2. 2016RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechanismcited by 344
  3. 2019On Mutual Information Maximization for Representation Learningcited by 315
  4. 2021Provable Guarantees for Self-Supervised Deep Learning with Spectral Contrastive Losscited by 311
  5. 2020CLUB: A Contrastive Log-ratio Upper Bound of Mutual Informationcited by 233
  6. 2021Contrastive Learning Inverts the Data Generating Processcited by 220
  7. 2015Simultaneous Deep Transfer Across Domains and Taskscited by 205
  8. 2020Estimating Counterfactual Treatment Outcomes over Time Through Adversarially Balanced Representationscited by 154
  9. 2020Contrastive learning, multi-view redundancy, and linear modelscited by 146
  10. 2022Causal Transformer for Estimating Counterfactual Outcomescited by 121
  11. 2020What makes instance discrimination good for transfer learning?cited by 111
  12. 2020Causal Shapley Values: Exploiting Causal Knowledge to Explain Individual Predictions of Complex Modelscited by 107
  13. 2018The Blessings of Multiple Causescited by 103
  14. 2022Understanding Contrastive Learning Requires Incorporating Inductive Biasescited by 96
  15. 2018Tighter Variational Bounds are Not Necessarily Bettercited by 76
  16. 2019Time Series Deconfounder: Estimating Treatment Effects over Time in the Presence of Hidden Confounderscited by 74
  17. 2018The Unusual Effectiveness of Averaging in GAN Trainingcited by 73
  18. 2023Factorized Contrastive Learning: Going Beyond Multi-view Redundancycited by 73
  19. 2023To Compress or Not to Compress- Self-Supervised Learning and Information Theory: A Reviewcited by 72
  20. 2022Continuous-Time Modeling of Counterfactual Outcomes Using Neural Controlled Differential Equationscited by 59
  21. 2023Identifiability Results for Multimodal Contrastive Learningcited by 58
  22. 2020Identifying Causal-Effect Inference Failure with Uncertainty-Aware Modelscited by 57
  23. 2021Conformal Sensitivity Analysis for Individual Treatment Effectscited by 40
  24. 2017Learning Independent Features with Adversarial Nets for Non-linear ICAcited by 33
  25. 2017Estimation of causal effects with multiple treatments: a review and new ideascited by 29
  26. 2022Estimating average causal effects from patient trajectoriescited by 29
  27. 2022Personalized Decision Making -- A Conceptual Introductioncited by 29
  28. 2021Sequential Deconfounding for Causal Inference with Unobserved Confounderscited by 24
  29. 2022Predicting the impact of treatments over time with uncertainty aware neural differential equationscited by 23
  30. 2023Counterfactual Generative Models for Time-Varying Treatmentscited by 17
  31. 2021Deconfounding Temporal Autoencoder: Estimating Treatment Effects over Time Using Noisy Proxiescited by 16
  32. 2023CF-GODE: Continuous-Time Causal Inference for Multi-Agent Dynamical Systemscited by 16
  33. 2018Multiple Causal Inference with Latent Confoundingcited by 14
  34. 2017Scalable Joint Models for Reliable Uncertainty-Aware Event Predictioncited by 13
  35. 2023Towards a Unified Framework of Contrastive Learning for Disentangled Representationscited by 13
  36. 2021Causal Mediation Analysis with Hidden Confounderscited by 10
  37. 2021Disentangled Counterfactual Recurrent Networks for Treatment Effect Inference over Timecited by 9
  38. 2022Learning Infomax and Domain-Independent Representations for Causal Effect Inference with Real-World Datacited by 9

50 most cited of 52 All 52 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 3; 2026: 4.

7 papers.

  1. 2025Transformer-Based Spatial-Temporal Counterfactual Outcomes Estimationcited by 3
  2. 2026CAETC: Causal Autoencoding and Treatment Conditioning for Counterfactual Estimation over Timecited by 1
  3. 2026Beyond Patient Invariance: Learning Cardiac Dynamics via Action-Conditioned JEPAscited by 1
  4. 2025Contrastive Representation Learning Helps Cross-institutional Knowledge Transfer: A Study in Pediatric Ventilation Managementcited by 0
  5. 2025Inferring Effects of Major Events through Discontinuity Forecasting of Population Anxietycited by 0
  6. 2026From Static Risk to Dynamic Trajectories: Toward World-Model-Inspired Clinical Predictioncited by 0
  7. 2026Causal State-Space Model for Causal Inference: Estimating Longitudinal Individual Treatment Effectscited by 0