Partially Observable Reinforcement Learning for Sustainable Active Surveillance.
Hechang ChenBo YangJiming LiuPublished in: KSEM (2) (2018)
Keyphrases
- partially observable
- reinforcement learning
- state space
- markov decision processes
- partial observability
- partially observable environments
- partially observable domains
- decision problems
- hidden state
- dynamical systems
- function approximation
- markov decision problems
- partial observations
- infinite horizon
- model free
- reinforcement learning algorithms
- reward function
- temporal difference
- belief space
- continuous state
- belief state
- action models
- machine learning
- optimal policy
- partially observable markov decision processes
- hidden markov models
- action space
- learning agent
- multi agent
- fully observable
- finite state
- objective function
- dynamic programming