BriefGPT.xyz
Mar, 2020
具有多样化上下文的随机线性情境策略带
Stochastic Linear Contextual Bandits with Diverse Contexts
HTML
PDF
Weiqiang Wu, Jing Yang, Cong Shen
TL;DR
本文研究了上下文多样性对随机线性情境赌博机的影响，提出了LinUCB-d算法并分析其遗憾性能，理论结果表明，在多样性上下文的假设下，LinUCB-d的期望累积遗憾被一个常数限制，改善了以往对LinUCB的理解并加强了其性能保证。
Abstract
In this paper, we investigate the impact of
context diversity
on
stochastic linear contextual bandits
. As opposed to the previous view that contexts lead to more difficult bandit learning, we show that when the c
→