BriefGPT.xyz
Sep, 2018
组合多臂赌博机中 Thompson Sampling 的分析与概率触发武器
Thompson Sampling for Combinatorial Multi-armed Bandit with Probabilistically Triggered Arms
HTML
PDF
Alihan Hüyük, Cem Tekin
TL;DR
研究了在半盲反馈条件下，组合多臂赌博问题中，具有概率触发武器的组合汤普森抽样的遗憾，并在基准武器预期的连续Lipschitz情况下得出了CTS的遗憾界。
Abstract
We analyze the
regret
of
combinatorial thompson sampling
(CTS) for the combinatorial
multi-armed bandit
with probabilistically triggered a
→