Causal bayesian optimization improves intervention decision making in complex systems
Causal Bayesian Optimization: Foundations, Methods, and Applications
Machine Learning
Summary
When you want to figure out the best action to change a system, understanding cause and effect helps you make better choices efficiently. The authors review a technique called causal Bayesian optimization that uses knowledge about how parts of a system affect each other to decide which interventions to try next. They compare different methods and test them across many scenarios, finding that no single approach works best in every case and that mistakes in understanding the system can affect results. They also highlight challenges like handling unknown systems and mixing different kinds of changes in future work.
What this means in practice
- •For healthcare data teams: Select efficient treatment plans by leveraging causal knowledge to reduce the number of clinical trial interventions.
- •For manufacturing engineers: Optimize process adjustments by systematically choosing interventions informed by causal system models to improve quality control.
A survey. It maps existing work.
Authors
Chenfeng Huang, Thuy T. Le, Zixuan Ma, Hien Tran
Abstract
Causal Bayesian Optimization (CBO) combines causal inference with Bayesian optimization to enable sample-efficient intervention selection in systems with causal structure. This survey provides a systematic review of CBO through a unified BO-loop perspective, showing how causal assumptions shape intervention search spaces, surrogate models, acquisition functions, and decision policies. We organize existing methods by graph and system-knowledge assumptions, environment, intervention representation, surrogate architecture, and decision rule, and connect CBO to causal bandits, Bayesian experimental design, safe optimization, policy search, and causal abstraction. We also introduce a reproducibility-oriented benchmark spanning hard- and soft-intervention settings, with standardized GAP and a new trajectory-aware Path-Aware GAP (PA-GAP), evaluating seven CBO methods and a non-causal BO baseline across thirteen datasets, three budgets, and two metrics. Results show that no method dominates uniformly: rankings depend on dataset, budget, metric, and how causal information is used, while strong non-causal baselines remain competitive in several settings. Controlled graph-misspecification and omitted-variable stress tests further show that rankings can change substantially when learner-side causal information is perturbed. We conclude by identifying key open challenges, including robustness to causal-assumption violations, scalable unknown-graph optimization, mixed intervention types, realistic cost models, stronger theoretical guarantees, and integration with modern representation learning and causal abstractions.