Constrained Causal Bayesian Optimization

TL;DR

This paper introduces constrained causal Bayesian optimization (cCBO), a method for efficiently identifying interventions in causal graphs that optimize a target while satisfying constraints, leveraging graph structure and data.

Contribution

The paper presents cCBO, a novel approach combining Gaussian process modeling and constrained optimization to improve intervention selection in causal graphs.

Findings

cCBO achieves fast convergence to feasible interventions.

It effectively balances optimization and constraint satisfaction.

Demonstrated success on artificial and real-world causal graphs.

Abstract

We propose constrained causal Bayesian optimization (cCBO), an approach for finding interventions in a known causal graph that optimize a target variable under some constraints. cCBO first reduces the search space by exploiting the graph structure and, if available, an observational dataset; and then solves the restricted optimization problem by modelling target and constraint quantities using Gaussian processes and by sequentially selecting interventions via a constrained expected improvement acquisition function. We propose different surrogate models that enable to integrate observational and interventional data while capturing correlation among effects with increasing levels of sophistication. We evaluate cCBO on artificial and real-world causal graphs showing successful trade off between fast convergence and percentage of feasible interventions.