A Generalized Reduced Linear Program for Markov Decision Processes.
Chandrashekar LakshminarayananShalabh BhatnagarPublished in: AAAI (2015)
Keyphrases
- markov decision processes
- linear program
- dynamic programming
- linear programming
- average cost
- approximate dynamic programming
- policy iteration
- semi infinite
- state space
- stationary policies
- finite state
- optimal policy
- transition matrices
- factored mdps
- reinforcement learning
- finite horizon
- decision theoretic planning
- optimal solution
- stochastic programming
- infinite horizon
- objective function
- action space
- partially observable
- model based reinforcement learning
- markov decision process
- np hard
- action sets
- reachability analysis
- column generation
- optimal control
- average reward
- planning under uncertainty
- markov decision problems
- multistage
- total cost
- reward function