Causal Dynamic Variational Autoencoder for Counterfactual Regression in Longitudinal Data
Not on the map. Why?
Abstract (arXiv)
Accurately estimating treatment effects over time is crucial in fields such as precision medicine, epidemiology, economics, and marketing. Many current methods for estimating treatment effects over time assume that all confounders are observed or attempt to infer unobserved ones. In contrast, our approach focuses on unobserved adjustment variables, which specifically have a causal effect on the outcome sequence. Under the assumption of unconfoundedness, we address estimating Conditional Average Treatment Effects (CATEs) while accounting for unobserved heterogeneity in response to treatment due to these unobserved adjustment variables. Our proposed Causal Dynamic Variational Autoencoder (CDVAE) is grounded in theoretical guarantees concerning the validity of latent adjustment variables and generalization bounds on CATE estimation error. Extensive evaluations on synthetic and real-world datasets show that CDVAE outperforms existing baselines. Moreover, we demonstrate that state-of-the-art models significantly improve their CATE estimates when augmented with the latent substitutes learned by CDVAE, approaching oracle-level performance without direct access to the true adjustment variables.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 57 of its 144 references that are in the map. First posted on arXiv 16 Oct 2023.
Builds on
57 of 144 references.
References · 57
- 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
- 2014Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translationcited by 4,057
- 2014Sequence to Sequence Learning with Neural Networkscited by 2,986
- 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,739
- 2019Optuna: A Next-generation Hyperparameter Optimization Frameworkcited by 2,501
- 2018Averaging Weights Leads to Wider Optima and Better Generalizationcited by 1,209
- 2015Generating Sentences from a Continuous Spacecited by 612
- 2015Estimation and Inference of Heterogeneous Treatment Effects using Random Forestscited by 600
- 2017Causal Effect Inference with Deep Latent-Variable Modelscited by 410
- 2016RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechanismcited by 349
- 2017Quasi-Oracle Estimation of Heterogeneous Treatment Effectscited by 332
- 2016Learning Representations for Counterfactual Inferencecited by 329
38 more
- 2016Ladder Variational Autoencoderscited by 318
- 2016Generalized Random Forestscited by 312
- 2019Adapting Neural Networks for the Estimation of Treatment Effectscited by 287
- 2016Variational Deep Embedding: An Unsupervised and Generative Approach to Clusteringcited by 220
- 2013Fast Computation of Wasserstein Barycenterscited by 216
- 2019Diagnosing and Enhancing VAE Modelscited by 193
- 2019Cyclical Annealing Schedule: A Simple Approach to Mitigating KL Vanishingcited by 191
- 2015Deep Kalman Filterscited by 166
- 2020Estimating Counterfactual Treatment Outcomes over Time Through Adversarially Balanced Representationscited by 155
- 2016Identifying Causal Effects With Proxy Variables of an Unmeasured Confoundercited by 139
- 2020Dynamical Variational Autoencoders: A Comprehensive Reviewcited by 128
- 2022Causal Transformer for Estimating Counterfactual Outcomescited by 123
- 2019Don't Blame the ELBO! A Linear VAE Perspective on Posterior Collapsecited by 121
- 2018Disentangled Sequential Autoencodercited by 120
- 2017Automated versus do-it-yourself methods for causal inference: Lessons learned from a data analysis competitioncited by 106
- 2018The Blessings of Multiple Causescited by 104
- 2019Support and Invertibility in Domain-Invariant Representationscited by 87
- 2021A Review of the Gumbel-max Trick and its Extensions for Discrete Stochasticity in Machine Learningcited by 82
- 2017REBAR: Low-variance, unbiased gradient estimates for discrete latent variable modelscited by 79
- 2019Time Series Deconfounder: Estimating Treatment Effects over Time in the Presence of Hidden Confounderscited by 75
- 2020Generalization Bounds and Representation Learning for Estimation of Potential Outcomes and Causal Effectscited by 72
- 2016Balancing Covariates via Propensity Score Weightingcited by 65
- 2014Learning Stochastic Recurrent Networkscited by 60
- 2023Posterior Collapse and Latent Variable Non-identifiabilitycited by 57
- 2020Counterfactual Representation Learning with Balancing Weightscited by 45
- 2016Double/Debiased Machine Learning for Treatment and Causal Parameterscited by 43
- 2018DeepMatch: Balancing Deep Covariate Representations for Causal Inference Using Adversarial Trainingcited by 42
- 2018Learning Weighted Representations for Generalization Across Designscited by 41
- 2022Personalized Decision Making -- A Conceptual Introductioncited by 30
- 2017Estimation of causal effects with multiple treatments: a review and new ideascited by 29
- 2018A comparison of methods for model selection when estimating individual treatment effectscited by 28
- 2021Multi-Facet Clustering Variational Autoencoderscited by 28
- 2020Double/Debiased Machine Learning for Dynamic Treatment Effects via g-Estimationcited by 27
- 2021Sequential Deconfounding for Causal Inference with Unobserved Confounderscited by 24
- 2019Comment on "Blessings of Multiple Causes"cited by 16
- 2015Structural Nested Models and G-estimation: The Partially Realized Promisecited by 15
- 2021Preventing Oversmoothing in VAE via Generalized Variance Parameterizationcited by 15
- 2018Multiple Causal Inference with Latent Confoundingcited by 14
50 most cited of 57 All 57 (JSON)
Built on by
Citing papers per year: 2026: 1.
1 paper.