Abstract
Policy gradient methods have become one of the most popular classes of algorithms for multi-agent reinforcement learning. A key challenge, however, that is not addressed by many of these methods is multi-agent credit assignment: assessing an agent’s contribution to the overall performance, which is crucial for learning good policies. We propose a novel algorithm called Dr.Reinforce that explicitly tackles this by combining difference rewards with policy gradients to allow for learning decentralized policies when the reward function is known. By differencing the reward function directly, Dr.Reinforce avoids difficulties associated with learning the 푄-function as done by Counterfactual Multiagent Policy Gradients (COMA), a state-of-the-art difference rewards method. For applications where the reward function is unknown, we show the effectiveness of a version of Dr.Reinforce that learns a reward network that is used to estimate the difference rewards.
Original language | English |
---|---|
Title of host publication | Proceedings of the 20th International Conference on Autonomous Agents and MultiAgent Systems |
Place of Publication | Richland, SC |
Publisher | International Foundation for Autonomous Agents and Multiagent Systems |
Pages | 1463-1465 |
Number of pages | 3 |
ISBN (Electronic) | 9781450383073 |
Publication status | Published - 2021 |
Event | 20th International Conference on Autonomous Agentsand Multiagent Systems - Virtual/online event due to COVID-19 Duration: 3 May 2021 → 7 May 2021 Conference number: 20 |
Publication series
Name | AAMAS '21 |
---|---|
Publisher | International Foundation for Autonomous Agents and Multiagent Systems |
ISSN (Electronic) | 2523-5699 |
Conference
Conference | 20th International Conference on Autonomous Agentsand Multiagent Systems |
---|---|
Abbreviated title | AAMAS 2021 |
Period | 3/05/21 → 7/05/21 |
Keywords
- Multi-Agent Reinforcement Learning
- Policy Gradients
- Difference Rewards
- Multi-Agent Credit Assignment
- Reward Learning
Fingerprint
Dive into the research topics of 'Difference Rewards Policy Gradients'. Together they form a unique fingerprint.Prizes
-
Best Paper Award
Castellini, Jacopo (Recipient), Devlin, Sam (Recipient), Oliehoek, F.A. (Recipient) & Savani, Rahul (Recipient), 4 May 2021
Prize: Prize (including medals and awards)