arXiv cs.MA· Arun Vignesh Malarkkan, Wangyang Ying, Yanjie Fu·· 12 小时前AI 评分31
CAFE：用因果引导与多智能体强化学习实现自动化特征工程
CAFE: Causally-Guided Automated Feature Engineering with Multi-Agent Reinforcement Learning
AI 导读
CAFE 将自动化特征工程重构为因果引导的序贯决策过程，先用因果发现得到软因果先验，再用级联多智能体深度 Q-learning 选择因果组与变换算子。在 15 个公开基准上，CAFE 相对最强 AFE 基线取得统计显著提升（Wilcoxon p<0.001，大效应），并减少收敛所需 episode 数。
正文
Abstract:Automated feature engineering (AFE) enables AI systems to autonomously construct high-utility representations from raw tabular data. However, existing AFE methods rely on statistical heuristics, yielding brittle features that fail under distribution shift. We introduce CAFE, a framework that reformulates AFE as a causally-guided sequential decision process, bridging causal discovery with reinforcement learning-driven feature construction. Phase I learns a sparse directed acyclic graph over features and the target to obtain soft causal priors, grouping features as direct, indirect, or other based on their causal influence with respect to the target. Phase II uses a cascading multi-agent deep Q-learning architecture to select causal groups and transformation operators, with hierarchical reward shaping and causal group-level exploration strategies that favor causally plausible transformations while controlling feature complexity. Across 15 public benchmarks (classification with macro-F1; regression with inverse relative absolute error), CAFE achieves statistically significant gains over the strongest AFE baseline (Wilcoxon p<0.001, large effect), reduces episodes-to-convergence, and delivers competitive time-to-target. Under both controlled covariate shifts and naturally occurring distribution shifts (Folktables), CAFE reduces out-of-distribution performance drop by up to 4x relative to a non-causal multi-agent baseline, and produces more compact feature sets with more stable post-hoc attributions. These findings underscore that causal structure, used as a soft inductive prior rather than a rigid constraint, can substantially improve the robustness and efficiency of automated feature engineering.
| Comments: | 10 Pages, References, and Appendix. Accepted at NeurIPS 2026 |
| Subjects: | Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Multiagent Systems (cs.MA) |
| ACM classes: | I.2.4; I.2.6; I.2.8; I.2.11 |
| Cite as: | arXiv:2602.16435 [cs.AI] |
| (or arXiv:2602.16435v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2602.16435 arXiv-issued DOI via DataCite |
Submission history
From: Arun Vignesh Malarkkan [view email]
[v1]
Wed, 18 Feb 2026 13:12:11 UTC (579 KB)
[v2]
Sun, 25 Oct 2026 14:58:58 UTC (599 KB)
来源：arXiv cs.MA · arxiv.org