Near-Optimal Learning and Planning in Separated Latent MDPs.

Fan Chen Constantinos Daskalakis Noah Golowich Alexander Rakhlin

Published in: COLT (2024)

Keyphrases

reinforcement learning
learning process
learning systems
stochastic domains
dynamic programming
online learning
learning algorithm
domain independent
prior knowledge
active learning
state space
learning tasks
decision theoretic
action selection
partially observable
structural svm
macro actions