EFXPLORER：基于 Epigraph 形式引导探索的多智能体协作强化学习方法
Cooperative Multi-Agent Reinforcement Learning via Epigraph-Form Guided Exploration
AI 导读
研究者提出 EFXPLORER，一种约束式探索框架，在保持任务性能的约束下最大化探索目标，通过 epigraph 重构引入自适应探索预算，将内在奖励与任务目标分离，并结合基于 successor distance 的内在奖励捕捉长时程依赖。该方法在 SMAX、VMAS 和 MPE 基准套件上超越强基线，诱导出新的协作策略。论文已被 NeurIPS 2026 接收。
正文 · 原文
Abstract:Discovering cooperative behavior in multi-agent reinforcement learning (MARL) is challenging due to the combinatorial complexity of joint state-action spaces, which hinders the emergence of coordinated behaviors from trial-and-error alone. Intrinsic rewards are often used to aid discovery, but naively combining them with team objectives can distort the learning signal, compromising task performance. In this paper, we propose EFXPLORER, a constrained exploration framework that maximizes exploration objectives subject to a constraint that preserves established task performance. We solve the resulting problem using an epigraph reformulation that introduces adaptive exploration budgets. This approach separates intrinsic rewards from task objectives and regulates exploration through task feasibility. To further encourage diverse and temporally extended exploration, we incorporate a successor distance-based intrinsic reward that captures long-horizon dependencies. Empirically, our method outperforms strong baselines and induces novel cooperative strategies across SMAX, VMAS, and MPE benchmark suites.
| Comments: | NeurIPS 2026 |
| Subjects: | Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2611.07066 [cs.AI] |
| (or arXiv:2611.07066v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.07066 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Hoseong Jung [view email]
[v1]
Wed, 4 Nov 2026 19:24:25 UTC (3,582 KB)
来源：arXiv cs.AI · arxiv.org