Beyond the One-Step Greedy Approach in Reinforcement Learning.

Yonathan Efroni Gal Dalal Bruno Scherrer Shie Mannor

Published in: ICML (2018)

Keyphrases

reinforcement learning
greedy algorithm
dynamic programming
post processing
learning algorithm
search algorithm
reinforcement learning algorithms
neural network
optimal policy
preprocessing
preprocessing step
learning problems
lower bound
feature selection
information systems
search engine
machine learning