Exploration and Communication for Partially Observable Collaborative Multi-Agent Reinforcement Learning.
Raphaël AvalosPublished in: AAMAS (2022)
Keyphrases
- multi agent reinforcement learning
- partially observable
- reinforcement learning
- distributed control
- state space
- markov decision processes
- decision problems
- dynamical systems
- multi agent
- learning agents
- belief state
- multi agent learning
- multi agent systems
- reward function
- infinite horizon
- cooperative
- action selection
- optimal policy
- domain specific
- learning process
- partially observable markov decision processes
- machine learning
- temporal difference
- learning tasks
- transfer learning
- random walk