Recurrent prediction model for partially observable MDPs.
Shaorong XieZhenyu ZhangHang YuXiangfeng LuoPublished in: Inf. Sci. (2023)
Keyphrases
- prediction model
- partially observable
- markov decision processes
- markov decision problems
- state space
- reinforcement learning
- decision problems
- dynamical systems
- regression model
- infinite horizon
- partial observability
- optimal policy
- partial observations
- belief state
- reward function
- neural network
- policy iteration
- finite state
- probabilistic planning
- partially observable markov decision process
- dec pomdps
- partially observable environments
- average reward
- markov decision process
- action models
- dynamic programming
- multi agent
- bayesian networks
- average cost
- planning under uncertainty
- planning domains
- optimal control
- fully observable
- knowledge base
- learning algorithm