From Tokens to Policy: Causal and Interpretable Heterogeneous Treatment Effects Identification
2026-06-15 • Machine Learning
Machine Learning
AI summaryⓘ
The authors focus on understanding how different people respond differently to treatments, especially when some important factors are not directly measured. They propose a new method called NEXIS that uses lots of pre-treatment data, including satellite images, to better identify hidden influences affecting treatment outcomes. Their approach treats this as a problem of finding the right set of factors (Markov blanket) from complex data representations. They tested NEXIS on real poverty-alleviation programs in Africa and found new insights to improve these programs.
Heterogeneous Treatment EffectCausal inferenceMarkov blanketLatent variablesMulti-modal dataRepresentation learningControlled experimentsSatellite imageryPolicy optimizationNeural networks
Authors
Riccardo Cadei, Frank Otchere, Nyasha Tirivayi, Gustavo Angeles Tagliaferro, Falco J. Bargagli-Stoffi, Francesco Locatello
Abstract
Heterogeneous Treatment Effect (HTE) identification is crucial to explain the impact of an intervention and optimize our policies accordingly. Existing approaches trade expressivity for interpretability, but, if some active heterogeneity drivers are unmeasured, methods at both ends of this spectrum allow for spurious HTE characterization with no causal reading. In this work, we focus on controlled experiments and argue that an oracle HTE causal characterization via the latent interactors is now within reach, thanks to (i) more extensive pre-treatment measurements, i.e., multi-modal and multi-view, and (ii) scalable representations with minimal human supervision. We then re-frame HTE identification as a Markov-blanket discovery problem on a sufficient and aligned pre-treatment representation, and introduce Neural EXposure Interaction Search (NEXIS), an iterative procedure with provable and empirically validated consistent selection. We deploy NEXIS on two anti-poverty programs in Africa, augmenting each with satellite imagery capturing previously unmeasured environmental effect modifiers, leading to novel, interpretable and prescriptive guidelines to optimize the programs' next iterations.