发表机构
University of California, Los Angeles; California State University, Long Beach; North Carolina State University（加州大学洛杉矶分校; 加州州立大学长滩分校; 北卡罗来纳州立大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文系统综述因果贝叶斯优化，提出统一贝叶斯优化循环视角，引入含新指标PA-GAP的可复现基准，实验表明无方法普遍占优，并指出鲁棒性等开放挑战。
AI 中文摘要
因果贝叶斯优化（CBO）将因果推断与贝叶斯优化相结合，以在具有因果结构的系统中实现样本高效的干预选择。本综述通过统一的贝叶斯优化循环视角对CBO进行系统性回顾，展示因果假设如何塑造干预搜索空间、代理模型、采集函数和决策策略。我们根据图与系统知识假设、环境、干预表示、代理架构和决策规则对现有方法进行归类，并将CBO与因果老虎机、贝叶斯实验设计、安全优化、策略搜索和因果抽象联系起来。我们还引入了一个面向可复现性的基准，涵盖硬干预和软干预设置，采用标准化的GAP和新的轨迹感知路径感知GAP（PA-GAP），在十三个数据集、三种预算和两种指标上评估了七种CBO方法和一种非因果贝叶斯优化基线。结果表明，没有一种方法在所有情况下都占优：排名取决于数据集、预算、指标以及因果信息的使用方式，而强大的非因果基线在若干设置中仍具有竞争力。受控的图错误设定和遗漏变量压力测试进一步表明，当学习者侧的因果信息受到扰动时，排名可能发生显著变化。最后，我们指出了关键开放挑战，包括对因果假设违反的鲁棒性、可扩展的未知图优化、混合干预类型、现实成本模型、更强的理论保证，以及与现代表示学习和因果抽象的集成。
英文摘要
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.
CommentsAccepted at Transactions on Machine Learning Research (TMLR), 2026
Journal refTransactions on Machine Learning Research, 2026
