Active Invariant Causal Prediction: Experiment Selection through Stability

TL;DR

This paper introduces A-ICP, an active learning framework that strategically selects experiments to efficiently identify causal relationships by leveraging invariance and stability concepts, reducing the number of costly interventions needed.

Contribution

It develops a new experiment selection method based on invariant causal prediction, improving causal discovery efficiency with theoretical and empirical validation.

Findings

A-ICP effectively identifies direct causes with fewer experiments.

Proposed policies outperform baseline methods in experiment efficiency.

Maintains error control while optimizing intervention selection.

Abstract

A fundamental difficulty of causal learning is that causal models can generally not be fully identified based on observational data only. Interventional data, that is, data originating from different experimental environments, improves identifiability. However, the improvement depends critically on the target and nature of the interventions carried out in each experiment. Since in real applications experiments tend to be costly, there is a need to perform the right interventions such that as few as possible are required. In this work we propose a new active learning (i.e. experiment selection) framework (A-ICP) based on Invariant Causal Prediction (ICP) (Peters et al., 2016). For general structural causal models, we characterize the effect of interventions on so-called stable sets, a notion introduced by (Pfister et al., 2019). We leverage these results to propose several intervention selection policies for A-ICP which quickly reveal the direct causes of a response variable in the causal graph while maintaining the error control inherent in ICP. Empirically, we analyze the performance of the proposed policies in both population and finite-regime experiments.