MuDE: Multi-agent decomposed reward-based exploration.
Byunghyun YooSungwon YiHyunwoo KimYounghwan ShinRan HanSeungwoo SeoHwa Jeon SongEuisok ChungJeongmin YangPublished in: Neural Networks (2024)
Keyphrases
- multi agent
- reinforcement learning
- cooperative
- exploration strategy
- multi agent systems
- cooperative agents
- single agent
- autonomous agents
- long run
- intelligent agents
- multiagent systems
- heterogeneous agents
- coalition formation
- learning algorithm
- databases
- multi agent environments
- active exploration
- reinforcement learning agents
- interactive exploration
- policy gradient
- temporal analysis
- action selection
- lower level
- multiscale
- real world