Proposes max-plus algebra framework to approximate optimal value functions in deterministic MDPs via fixed dictionaries (complexity tied to covering numbers) and adaptive matching-pursuit style bases (empirically effective but without guarantees).
Feature selection and feature learning for high- dimensional batch reinforcement learning: A survey.International Journal of Automation and Computing, 12(3):229–242, 2015
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Max-Plus Matching Pursuit for Deterministic Markov Decision Processes
Proposes max-plus algebra framework to approximate optimal value functions in deterministic MDPs via fixed dictionaries (complexity tied to covering numbers) and adaptive matching-pursuit style bases (empirically effective but without guarantees).