Causal Discovery by Interventions via Integer Programming

TL;DR

This paper introduces an integer programming approach for designing minimal intervention sets to reliably identify causal structures, overcoming limitations of observational methods and adaptable to various experimental constraints.

Contribution

It proposes a novel optimization-based method for causal discovery through interventions, providing exact, modular solutions tailored to different experimental settings.

Findings

Effective in identifying causal structures across diverse settings

Provides exact solutions for intervention design

Demonstrates robustness and applicability in comparative analyses

Abstract

Causal discovery is essential across various scientific fields to uncover causal structures within data. Traditional methods relying on observational data have limitations due to confounding variables. This paper presents an optimization-based approach using integer programming (IP) to design minimal intervention sets that ensure causal structure identifiability. Our method provides exact and modular solutions that can be adjusted to different experimental settings and constraints. We demonstrate its effectiveness through comparative analysis across different settings, demonstrating its applicability and robustness.