Periodic agent-state based Q-learning for POMDPs.
Amit SinhaMathieu GeistAditya MahajanPublished in: CoRR (2024)
Keyphrases
- state space
- reinforcement learning
- multi agent
- belief state
- partially observable
- state action
- action selection
- cooperative
- agent receives
- learning agent
- single agent
- dynamic programming
- partial observations
- markov decision process
- state abstraction
- multi agent reinforcement learning
- continuous state
- multi agent systems
- partially observable markov decision processes
- reinforcement learning algorithms
- markov decision processes
- optimal policy
- intelligent agents
- autonomous agents
- linear programming
- action space
- average reward
- multiagent systems
- multiple agents
- state transition
- learning rate
- stochastic domains
- software agents
- discounted reward