The Representational Capacity of Action-Value Networks for Multi-Agent Reinforcement Learning
| Autor: | Jacopo Castellini, Frans Oliehoek, Rahul Savani, Shimon Whiteson |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2019 |
| Předmět: | |
| Zdroj: | 18th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2019: Proceedings of the Eighteenth International Conference on Autonomous Agents and Multiagent Systems (AAMAS) 18th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2019 Delft University of Technology |
| Popis: | Recent years have seen the application of deep reinforcement learning techniques to cooperative multi-agent systems, with great empirical success. However, given the lack of theoretical insight, it remains unclear what the employed neural networks are learning, or how we should enhance their representational power to address the problems on which they fail. In this work, we empirically investigate the representational power of various network architectures on a series of one-shot games. Despite their simplicity, these games capture many of the crucial problems that arise in the multi-agent setting, such as an exponential number of joint actions or the lack of an explicit coordination mechanism. Our results quantify how well various approaches can represent the requisite value functions, and help us identify issues that can impede good performance. This work as been accepted as an Extended Abstract in Proc. of the 18th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2019), N. Agmon, M. E. Taylor, E. Elkind, M. Veloso (eds.), May 2019, Montreal, Canada |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|