Skip to main content
Aggregate arXiv cs.AI 人工智能 18 Aug 2026 - 12:00

When to Communicate: Belief Distributions and KL Divergence for Principled Gating in Multi-Agent RL

RSS 官方收录 · 可信分层展示

关键摘要

arXiv:2608.…

  • 14559v1 Announce Type: new Abstract: Effective communication in multi-…
  • Existing approaches either communicate at every timestep or learn a bi…
  • I propose a principled alternative: agents communicate only when the K…

摘要引擎:抽取

正文提要

arXiv:2608.14559v1 Announce Type: new Abstract: Effective communication in multi-agent reinforcement learning requires agents to decide not only \textit{what} to communicate, but when? Existing approaches either communicate at every timestep or learn a binary gate through REINFORCE policy gradients \cite{singh2019}, a high-variance signal that produces unstable and uninterpretable gating behavior. I propose a principled alternative: agents communicate only when the KL divergence between their learned belief distributions exceeds a fixed threshold. Each agent maintains a belief distribution over a latent world state computed as a softmax over its LSTM hidden state, and communicates only when belief disagreement is large enough to justify information exchange. I evaluate this approach on the Predator-Prey benchmark from IC3Net \cite{singh2019} across two environment sizes with 5 seeds each, and on MPE simple\_spread \cite{lowe2017}, comparing against IC3Net, CommNet, and an independent controller. On PP 10$\times$10, IC3Net outperforms KL-belief at all thresholds. On the harder PP 20$\times$20, a threshold ablation over $\varepsilon \in \{0.1, 0.3, 0.5, 1.0\}$ reveals an inverted U-shape: $\varepsilon=0.5$ achieves 73.84 average steps and 42\% success rate versus IC3Net's 75.31 steps and 31\%, a gap of 1.47 steps and 11 percentage points with tighter seed variance. On MPE, the belief head improves mean reward by 12 points and reduces variance by 26$\times$ even when gating is inactive, suggesting two orthogonal contributions: principled gating when beliefs can converge, and improved latent representations that benefit coordination regardless.

来源:https://arxiv.org/abs/2608.14559

打开官方原文 站点原文页 可信分区 本信源更多 今日简报 分享图 RSS 稍后再看列表