arXiv cs.MA· Ali Atiah Alzahrani·· 13 小时前AI 评分31
Conservative Meta-Agent Search（CMAS）：非平稳环境下的智能体架构自适应搜索
Conservative Meta-Agent Search: Architecture Adaptation under Non-Stationarity
AI 导读
研究者提出 Conservative Meta-Agent Search（CMAS），将元智能体架构搜索形式化为带切换成本和硬性风险约束的序贯替换问题，通过配对影子回放与干预式组件归因打分，晋升需超过预设边际并经过模拟金丝雀阶段与回滚。
正文 · 原文
Abstract:Automated agentic-system design is typically evaluated on static task suites with immediate scalar rewards. Deployed systems face persistent shift, delayed dependent feedback, correlated component failures, and costly architecture changes. We formalize meta-agent architecture search as sequential incumbent replacement with switching costs and hard risk constraints and introduce Conservative Meta-Agent Search (CMAS). CMAS retrieves archived architectures and proposes typed graph edits, then scores each against the incumbent by paired shadow replay under common random numbers, with interventional component credit. Promotion requires an empirically calibrated, variance-penalized simultaneous score above a pre-specified margin, and is followed by a simulated canary stage and pre-specified rollback. Since backtests cannot certify meta-level skill, we introduce MetaInvest-Bench, a regime-switching simulator whose specified worker competence makes a surrogate-optimal architecture computable, with leakage-controlled replay, adversarial governance tracks, and five negative controls. Financial streams serve as a risk-sensitive testbed for meta-level generalization; we claim no alpha. Across 96 streams, CMAS cuts normalized mean surrogate dynamic-oracle regret from 0.116 to 0.104 and raises held-out utility from 0.44 to 0.48 over the strongest adaptive baseline; when extra workers add no value, it contracts to near-single-agent teams. Its deployment inference cost is ~6% lower, though its search-inclusive token cost is the highest of the methods compared. The observed per-update false-promotion rate is 3.4% against a nominal 5% target, and 4.8% in a sealed rerun. The calibration is empirical: it concerns an update-level replay statistic on an author-designed benchmark, and guarantees neither full-utility improvement, nor absence of simulated canary-stage harm, nor sequence-level error control.
| Comments: | 24 pages, 8 figures, 17 tables |
| Subjects: | Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2610.31436 [cs.AI] |
| (or arXiv:2610.31436v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.31436 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Ali Atiah Alzahrani Mr [view email]
[v1]
Thu, 22 Oct 2026 15:32:06 UTC (411 KB)
来源：arXiv cs.MA · arxiv.org