Anderson acceleration for partially observable Markov decision processes: A maximum entropy approach.
Mingyu ParkJaeuk ShinInsoon YangPublished in: Autom. (2024)
Keyphrases
- maximum entropy
- partially observable markov decision processes
- finite state
- dynamical systems
- reinforcement learning
- decision problems
- belief state
- maximum entropy principle
- dynamic programming
- optimal policy
- partially observable stochastic games
- markov decision processes
- state space
- multi agent
- planning problems
- transformation based learning
- markov models
- partially observable
- minimum cross entropy
- np hard
- infinite horizon
- linear programming
- knowledge base
- domain independent
- markov chain