A Reinforcement Learning Scheme for a Partially-Observable Multi-Agent Game.
Shin IshiiHajime FujitaMasaoki MitsutakeTatsuya YamazakiJun MatsudaYoichiro MatsunoPublished in: Mach. Learn. (2005)
Keyphrases
- learning scheme
- partially observable
- reinforcement learning
- multi agent
- learning algorithm
- state space
- markov decision processes
- partial observability
- markov decision problems
- partial observations
- game theory
- action models
- initially unknown
- machine learning
- decision problems
- nash equilibrium
- single agent
- multi agent systems
- optimal policy
- partially observable environments
- dynamical systems
- temporal difference
- reinforcement learning algorithms
- dynamic programming
- transfer learning
- special case
- coalition formation
- partially observable markov decision processes
- infinite horizon
- reward function
- orders of magnitude
- pairwise