Sleeping Experts and Bandits Approach to Constrained Markov Decision Processes

Hyeong Soo Chang

This brief paper presents simple simulation-based algorithms for obtaining an approximately optimal policy in a given finite set in large finite constrained Markov decision process…