Smoke and Mirrors in Causal Downstream Tasks

TL;DR

This paper investigates the biases in causal inference methods using high-dimensional data, introduces a real-world benchmark with ants, and provides guidelines for better representation learning in scientific causal tasks.

Contribution

It reveals biases in common causal inference choices, introduces ISTAnt, a novel real-world benchmark, and offers guidelines for representation learning in scientific causal analysis.

Findings

Sampling and modeling choices significantly impact causal estimate accuracy.

Classification accuracy does not reliably indicate causal estimate quality.

Real-world and synthetic experiments confirm the importance of careful causal modeling.

Abstract

Machine Learning and AI have the potential to transform data-driven scientific discovery, enabling accurate predictions for several scientific phenomena. As many scientific questions are inherently causal, this paper looks at the causal inference task of treatment effect estimation, where the outcome of interest is recorded in high-dimensional observations in a Randomized Controlled Trial (RCT). Despite being the simplest possible causal setting and a perfect fit for deep learning, we theoretically find that many common choices in the literature may lead to biased estimates. To test the practical impact of these considerations, we recorded ISTAnt, the first real-world benchmark for causal inference downstream tasks on high-dimensional observations as an RCT studying how garden ants (Lasius neglectus) respond to microparticles applied onto their colony members by hygienic grooming. Comparing 6 480 models fine-tuned from state-of-the-art visual backbones, we find that the sampling and modeling choices significantly affect the accuracy of the causal estimate, and that classification accuracy is not a proxy thereof. We further validated the analysis, repeating it on a synthetically generated visual data set controlling the causal model. Our results suggest that future benchmarks should carefully consider real downstream scientific questions, especially causal ones. Further, we highlight guidelines for representation learning methods to help answer causal questions in the sciences.