Improving Policies via Search in Cooperative Partially Observable Games.
Adam LererHengyuan HuJakob N. FoersterNoam BrownPublished in: CoRR (2019)
Keyphrases
- partially observable
- cooperative
- markov decision problems
- game theory
- markov decision processes
- state space
- search algorithm
- partially observable markov decision processes
- reward function
- dynamical systems
- decision problems
- search strategy
- reinforcement learning
- infinite horizon
- search strategies
- optimal policy
- search space
- partial observability
- partially observable environments
- partially observable domains
- belief state
- evaluation function
- search methods
- heuristic search
- multi agent systems