Structured World Belief for Reinforcement Learning in POMDP.
Gautam SinghSkand Vishwanath PeriJunghyun KimHyunseok KimSungjin AhnPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- belief state
- state space
- belief space
- partially observable markov decision processes
- partially observable
- continuous state
- hidden state
- partially observable markov decision process
- markov decision processes
- point based value iteration
- optimal policy
- function approximation
- model free
- temporal difference
- reinforcement learning algorithms
- partial observability
- markov decision process
- learning algorithm
- agent learns
- model free reinforcement learning
- machine learning
- multi agent
- dynamic programming
- state and action spaces
- learning process
- structured data
- policy evaluation
- belief functions
- markov decision problems
- reward function
- linear programming
- policy search
- mobile robot
- belief revision
- planning under uncertainty
- optimal control
- finite state