Fine-Grained Gap-Dependent Bounds for Tabular MDPs via Adaptive Multi-Step Bootstrap.

Haike Xu Tengyu Ma Simon S. Du

Published in: COLT (2021)

Keyphrases

fine grained
multi step
coarse grained
markov decision processes
lower bounding
access control
lower bound
upper bound
single step
knn
k nearest neighbor
contingency tables
tightly coupled
data lineage
reinforcement learning
neural network
data provenance
user intent
data sets