arXiv cs.AI· Catherine Ji, Vivek Myers, Joshua Lin, Alexandra Wingate, Sergey Levine, Benjamin Eysenbach·· 6 天前AI 评分31
Scaling Empowered Agents：从技能学习到赋权最大化的可扩展方法
Scaling Empowered Agents
AI 导读
研究者提出一种可扩展的赋权最大化方法，将赋权定义为某状态下可执行技能的数量，最大化赋权即抵达适合多种技能的起始状态，该方法适用于任意底层技能学习算法。实验显示赋权最大化涌现出资源采集、环境改造、生存等特性，并在 Crafter 中学会建造房屋。论文与代码已公开，共 30 页、13 张图。
正文 · 原文
Abstract:Empowerment measures the capacity of an agent to actively control its environment. In this work, we explain the connection between skill learning and empowerment. While skill learning is not empowerment, we introduce a simple scalable recipe for empowerment maximization: empowerment corresponds to the number of executable skills from a state, so maximizing empowerment corresponds to reaching states that are good starting points for many skills. Our recipe applies to arbitrary underlying skill learning methods. Whereas most prior work on empowerment looks at tabular settings, this recipe enables us to study empowerment at a scale where we can realize theoretically-predicted but empirically-elusive properties of empowerment (e.g., gathering resources, environment modification, survival). Experiments show that empowerment maximization unlocks interesting emergent properties, such as building a house in Crafter. Website and code at \href{this https URL}{this https url}.
| Comments: | 30 pages, 13 figures |
| Subjects: | Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.13002 [cs.AI] |
| (or arXiv:2610.13002v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.13002 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Catherine Ji [view email]
[v1]
Fri, 9 Oct 2026 05:04:30 UTC (5,635 KB)
来源：arXiv cs.AI · arxiv.org