Model-based Reinforcement Learning for Semi-Markov Decision Processes with Neural ODEs.

Jianzhun Du Joseph Futoma Finale Doshi-Velez

Published in: NeurIPS (2020)

Keyphrases

model based reinforcement learning
semi markov decision processes
markov decision processes
average reward
finite state
optimal policy
state space
reinforcement learning
dynamic programming
policy iteration
average cost
partially observable
markov decision process
decision processes
action space
infinite horizon