A Simple and Optimal Policy Design for Online Learning with Safety against Heavy-tailed Risk.

David Simchi-Levi Zeyu Zheng Feng Zhu

Published in: CoRR (2022)

Keyphrases

active learning
online learning
optimal policy
heavy tailed
state space
dynamic programming
markov decision processes
infinite horizon
finite horizon
multistage
develop a mathematical model
e learning
long run
least squares
finite state