Nearly Optimal Latent State Decoding in Block MDPs.

Yassir Jedra Junghyun Lee Alexandre Proutière Se-Young Yun

Published in: AISTATS (2023)

Keyphrases

state space
markov decision processes
reinforcement learning
dynamic programming
finite horizon
optimal solution
worst case
action sets
video sequences
optimal policy
state variables
action space
decoding algorithm