Model-Based Reinforcement Learning for Partially Observable Games with Sampling-Based State Estimation.
Hajime FujitaShin IshiiPublished in: Neural Comput. (2007)
Keyphrases
- state estimation
- partially observable
- model based reinforcement learning
- markov decision processes
- markov decision problems
- reinforcement learning
- state space
- particle filter
- kalman filter
- dynamic systems
- finite state
- kalman filtering
- optimal policy
- policy iteration
- decision problems
- infinite horizon
- particle filtering
- visual tracking
- dynamical systems
- dynamic programming
- partially observable markov decision processes
- decision processes
- reward function
- average cost
- action space
- belief state
- orders of magnitude
- state variables
- machine learning
- linear programming
- learning algorithm