2 papers
cs.MA2024
Augmenting the action space with conventions to improve multi-agent cooperation in Hanabi
F. Bredell, H. A. Engelbrecht, J. C. Schoeman
The card game Hanabi is considered a strong medium for the testing and development of multi-agent reinforcement learning (MARL) algorithms, due to its cooperative nature, partial o…
cs.LG2022
Credit-cognisant reinforcement learning for multi-agent cooperation
F. Bredell, H. A. Engelbrecht, J. C. Schoeman
Traditional multi-agent reinforcement learning (MARL) algorithms, such as independent Q-learning, struggle when presented with partially observable scenarios, and where agents are…