Weakly Coupled Constrained Markov Decision Processes in Borel Spaces.
Mukul GagraniAshutosh NayyarPublished in: ACC (2020)
Keyphrases
- markov decision processes
- finite state
- reinforcement learning
- optimal policy
- state space
- dynamic programming
- policy iteration
- finite horizon
- transition matrices
- decision processes
- planning under uncertainty
- model based reinforcement learning
- reinforcement learning algorithms
- reachability analysis
- decision theoretic planning
- average reward
- infinite horizon
- factored mdps
- stationary policies
- risk sensitive
- action space
- action sets
- markov decision process
- reward function
- partially observable markov decision processes
- state and action spaces
- state abstraction
- machine learning
- average cost