arXiv cs.MA· Muhammad Ahmed Atif, Nehal Naeem Haji, Muhammad Ebad Atif, Mohammad Shahid Shaikh·· 13 小时前AI 评分24
集中式学习真能帮具身智能体协作吗？一项表格型研究给出的边界条件
Does Centralized Learning Help Embodied Agents Coordinate, and When Can a Tabular Study Tell?
AI 导读
在捕食者-猎物环境中对比独立Q学习（IQL）与集中Q学习（CQL）的表格型研究发现，CQL在5x5和8x8网格、四种捕食者与猎物配对下均让智能体付出代价，其捕食者协作水平并不高于IQL。
正文 · 原文
Abstract:Tabular studies are a common way to compare independent and centralized multi-agent learners, because they avoid function approximation. A natural expectation is that centralized learning, which learns joint actions, helps agents coordinate, especially when their bodies make solo success hard. Earlier work tests this only in parts: it measures coordination for one learning method, compares methods with the agents' bodies fixed, or varies speed but reports only how well the team did. We test it in a predator-prey environment, where teammates cooperatively hunt together while two teams compete. We compare independent Q-learning (IQL) with central Q-learning (CQL) in all four predator and prey pairings, under limits on speed and stamina, on a 5x5 and an 8x8 grid trained for 200,000 episodes each, and check that the agents learn on both. We measure coordination directly, with predictions for the main measurement fixed in advance. In this task, where one predator can capture prey alone, we find three things. First, CQL costs its agents in both roles on both grids, and its predators coordinate no more than IQL's. Second, its one benefit comes from acting on the best joint action: this cut capture time by 42% on the small grid with faster prey and unlimited stamina, but helped nowhere on the larger grid, where the best joint action had mostly never been tried. Third, on the larger grid a tie-breaking default decided which method won, even though the agents had passed a learning check. A tabular comparison can therefore be trusted only if it also checks how much of the state space the agents covered and implements the methods alike. We propose simple checks for both.
| Comments: | 10 pages, 2 figures. Under peer review at AAMAS 2027 |
| Subjects: | Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2601.17454 [cs.MA] |
| (or arXiv:2601.17454v3 [cs.MA] for this version) | |
| https://doi.org/10.48550/arXiv.2601.17454 arXiv-issued DOI via DataCite |
Submission history
From: Muhammad Ahmed Atif [view email]
[v1]
Sat, 24 Jan 2026 13:09:01 UTC (919 KB)
[v2]
Fri, 24 Jul 2026 13:21:44 UTC (1,175 KB)
[v3]
Wed, 21 Oct 2026 17:09:52 UTC (149 KB)
来源：arXiv cs.MA · arxiv.org