Markov decision processes with maximum entropy rate for Surveillance Tasks.
Yu ChenShaoyuan LiXiang YinPublished in: CoRR (2022)
Keyphrases
- maximum entropy
- markov decision processes
- finite state
- maximum entropy principle
- reinforcement learning
- transition matrices
- markov models
- optimal policy
- state space
- decision theoretic planning
- policy iteration
- conditional random fields
- random fields
- dynamic programming
- probabilistic logic
- infinite horizon
- partially observable
- reachability analysis
- principle of maximum entropy
- average cost
- planning under uncertainty
- action space
- action sets
- model based reinforcement learning
- average reward
- markov decision process
- transformation based learning
- information retrieval
- generative model
- markov chain
- graphical models
- minimum cross entropy