Policy for Partially Observable Markov Decision Processes in Large Domains: Embedding Exploration Dynamics.
Giorgos ApostolikasSpyros G. TzafestasPublished in: Intell. Autom. Soft Comput. (2004)
Keyphrases
- partially observable markov decision processes
- dynamical systems
- partially observable domains
- sequential decision making problems
- optimal policy
- finite state
- reinforcement learning
- decision problems
- continuous state
- partial observability
- planning under uncertainty
- belief space
- belief state
- partially observable
- dynamic programming
- state space
- markov decision processes
- policy search
- partially observable environments
- partially observable stochastic games
- multi agent
- infinite horizon
- policy gradient
- inverse reinforcement learning
- fully observable
- partially observable markov decision process
- planning problems
- state dependent
- point based value iteration
- stochastic domains
- machine learning
- bayesian reinforcement learning
- action selection
- markov decision problems
- graphical models
- dec pomdps
- long run