mlsanityData to 14 Oct 2026

Image-based Treatment Effect Heterogeneity

arXiv 2206.06417 ↗First posted 13 Jun 2022cs.LG stat.MEcited by 20

Not on the map. Why?

Abstract (arXiv)

Randomized controlled trials (RCTs) are considered the gold standard for estimating the average treatment effect (ATE) of interventions. One use of RCTs is to study the causes of global poverty -- a subject explicitly cited in the 2019 Nobel Memorial Prize awarded to Duflo, Banerjee, and Kremer "for their experimental approach to alleviating global poverty." Because the ATE is a population summary, anti-poverty experiments often seek to unpack the effect variation around the ATE by conditioning (CATE) on tabular variables such as age and ethnicity that were measured during the RCT data collection. Although such variables are key to unpacking CATE, using only such variables may fail to capture historical, geographical, or neighborhood-specific contributors to effect variation, as tabular RCT data are often only observed near the time of the experiment. In global poverty research, when the location of the experiment units is approximately known, satellite imagery can provide a window into such factors important for understanding heterogeneity. However, there is no method that specifically enables applied researchers to analyze CATE from images. In this paper, using a deep probabilistic modeling framework, we develop such a method that estimates latent clusters of images by identifying images with similar treatment effects distributions. Our interpretable image CATE model also includes a sensitivity factor that quantifies the importance of image segments contributing to the effect cluster prediction. We compare the proposed methods against alternatives in simulation; also, we show how the model works in an actual RCT, estimating the effects of an anti-poverty intervention in northern Uganda and obtaining a posterior predictive distribution over effects for the rest of the country where no experimental data was collected. We make all models available in open-source software.

Key facts. Cited by 20 papers in the map (data through 14 Oct 2026). Builds on 31 of its 60 references that are in the map. First posted on arXiv 13 Jun 2022.

Builds on

31 of 60 references.

References · 31

  1. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
  2. 2019CLEVRER: CoLlision Events for Video REpresentation and Reasoningcited by 484
  3. 2017Meta-learners for Estimating Heterogeneous Treatment Effects using Machine Learningcited by 478
  4. 2017Causal Effect Inference with Deep Latent-Variable Modelscited by 410
  5. 2017Quasi-Oracle Estimation of Heterogeneous Treatment Effectscited by 332
  6. 2013Black Box Variational Inferencecited by 319
  7. 2016Generalized Random Forestscited by 312
  8. 2015Recursive Partitioning for Heterogeneous Causal Effectscited by 255
  9. 2020Deep Structural Causal Models for Tractable Counterfactual Inferencecited by 211
  10. 2021Towards Causal Representation Learningcited by 209
  11. 2019Causality matters in medical imagingcited by 193
  12. 2018Flipout: Efficient Pseudo-Independent Weight Perturbations on Mini-Batchescited by 133
19 more
  1. 2016Discovering Causal Signals in Imagescited by 79
  2. 2015External Validity: From Do-Calculus to Transportability Across Populationscited by 75
  3. 2020Using satellite imagery to understand and promote sustainable developmentcited by 71
  4. 2014Visual Causal Feature Learningcited by 59
  5. 2021Dynamic Visual Reasoning by Learning Differentiable Physics Models from Video and Languagecited by 55
  6. 2018DeepMatch: Balancing Deep Covariate Representations for Causal Inference Using Adversarial Trainingcited by 42
  7. 2021Causal Effect Inference for Structured Treatmentscited by 39
  8. 2019Specifying Weight Priors in Bayesian Deep Neural Networks with Empirical Bayescited by 32
  9. 2015Multi-Level Cause-Effect Systemscited by 30
  10. 2016Unsupervised Discovery of El Nino Using Causal Feature Learning on Microlevel Climate Datacited by 20
  11. 2021Regularizing towards Causal Invariance: Linear Models with Proxiescited by 20
  12. 2020Statistical modeling: the three culturescited by 18
  13. 2022Personalized Public Policy Analysis in Social Sciences using Causal-Graphical Normalizing Flowscited by 16
  14. 2017Selective inference for effect modification via the lassocited by 12
  15. 2021A unified view of likelihood ratio and reparameterization gradientscited by 12
  16. 2014Randomization Inference for Treatment Effect Variationcited by 9
  17. 2022Estimating Heterogeneous Causal Effects of High-Dimensional Treatments: Application to Conjoint Analysiscited by 7
  18. 2020Causal Inference in the Presence of Interference in Sponsored Search Advertisingcited by 6
  19. 2019Causal inference for climate change events from satellite image time series using computer vision and deep learningcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 4; 2024: 8; 2025: 7; 2026: 1.

20 papers.

Sort:

  1. 2024Causal Inference with Generative Artificial Intelligence: Application to Texts as Treatmentscited by 16
  2. 2023Causal Effect Estimation: Recent Advances, Challenges, and Opportunitiescited by 14
  3. 2024Modular Learning of Deep Causal Generative Models for High-dimensional Causal Inferencecited by 9
  4. 2025Prediction-Powered Causal Inferencescited by 9
  5. 2023Integrating Earth Observation Data into Causal Inference: Challenges and Opportunitiescited by 8
  6. 2024A Scoping Review of Earth Observation and Machine Learning for Causal Inference: Implications for the Geography of Povertycited by 5
  7. 2024Optimizing Multi-Scale Representations to Detect Effect Heterogeneity Using Earth Observation and Computer Vision: Applications to Two Anti-Poverty RCTscited by 4
  8. 2025Adjustment for Confounding using Pre-Trained Representationscited by 4
  9. 2023SSL Framework for Causal Inconsistency between Structures and Representationscited by 2
  10. 2024DISCRET: Synthesizing Faithful Explanations For Treatment Effect Estimationcited by 2
  11. 2025Debiasing Machine Learning Predictions for Causal Inference Without Additional Ground Truth Data: "One Map, Many Trials" in Satellite-Driven Poverty Analysiscited by 2
  12. 2023CausalImages: An R Package for Causal Inference with Earth Observation, Bio-medical, and Social Science Imagescited by 1
8 more
  1. 2025ReLU integral probability metric and its applicationscited by 1
  2. 2024Towards Causal Relationship in Indefinite Data: Baseline Model and New Datasetscited by 0
  3. 2024Enhancing predictive imaging biomarker discovery through treatment effect analysiscited by 0
  4. 2024Compositional Models for Estimating Causal Effectscited by 0
  5. 2025Annotation-Assisted Learning of Treatment Policies From Multimodal Electronic Health Recordscited by 0
  6. 2025Detecting and Mitigating Treatment Leakage in Text-Based Causal Inference: Distillation and Sensitivity Analysiscited by 0
  7. 2025Beyond Interaction Effects: Two Logics for Studying Population Inequalitiescited by 0
  8. 2026From Tokens to Policy: Causal and Interpretable Heterogeneous Treatment Effects Identificationcited by 0