arXiv cs.AI· Wenhan Wu, Linghan Chen, Yudong Gao, Fan Yang, Honglong Chen, Kai Yin, Jiawei Jiang, Chuang Hu·· 4 小时前AI 评分28
CARVE:多智能体强化学习中的语义协作策略遗忘方法
When Cooperation Must Change: Semantic Strategy Unlearning in Multi-Agent Reinforcement Learning
AI 导读
arXiv 论文提出 CARVE,将多智能体强化学习中删除特定协作约定的任务形式化为“语义协作策略遗忘”,用因果协作归因和保留感知子空间更新定位并清除目标约定,同时保留有效协作。在五个协作 MARL 基准上,该方法将约定完成率与最坏情况诱发率分别降低 91.4% 和 83.3%,中位效用损失为 5.9%,推理延迟与原策略持平。
正文 · 原文
Abstract:Cooperative multi-agent reinforcement learning (MARL) supports coordinated decision-making in distributed Web services, multi-robot control, and collaborative games through learned coordination conventions. When deployment requirements change, a previously useful convention may become obsolete or unsafe. For example, revised cache-sharing permissions in Web content delivery may require caches to abandon a learned division of storage and serving roles while maintaining efficient delivery. Teams must therefore remove the specified convention while preserving useful cooperation. This is challenging because the same relational pattern can be reproduced by different agents across different histories. Runtime shielding leaves the underlying coordination preference intact and can impair task progress, while existing reinforcement unlearning and trajectory-deletion methods do not directly target role-invariant conventions. This paper formalizes this problem as semantic coordination-strategy unlearning and proposes this http URL uses causal coordination attribution to identify influential interactions and retain-aware subspace updates to limit interference with unrelated behaviors. Counterfactual re-coordination helps agents learn productive alternatives, while adversarial auditing tests whether the target convention can still be recovered. Accepted updates are merged into decentralized actors for direct deployment. Evaluation across five cooperative MARL benchmarks shows that CARVE reduces mean nominal convention-completion and worst-case elicitation rates by 91.4\% and 83.3\%, this http URL median utility loss is 5.9\%, and measured inference latency matches that of the original policies.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.17917 [cs.AI] |
| (or arXiv:2611.17917v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.17917 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Wenhan Wu [view email]
[v1]
Fri, 13 Nov 2026 08:57:49 UTC (622 KB)
来源:arXiv cs.AI · arxiv.org