Symmetry in Markov Decision Processes and its Implications for Single Agent and Multiagent Learning.
Martin ZinkevichTucker R. BalchPublished in: ICML (2001)
Keyphrases
- multiagent learning
- markov decision processes
- single agent
- multi agent
- reinforcement learning
- action space
- multiple agents
- decision problems
- optimal policy
- dynamic environments
- state space
- finite state
- multi agent systems
- policy iteration
- learning agents
- multiagent systems
- dynamic programming
- planning under uncertainty
- dec pomdps
- path finding
- average reward
- stochastic games
- reinforcement learning algorithms
- infinite horizon
- partially observable
- policy gradient
- markov decision process
- cooperative
- autonomous agents
- decision making
- continuous state
- resource allocation
- learning algorithm
- search space
- mobile robot
- utility function
- average cost
- dynamical systems
- game theoretic