BriefGPT.xyz
Sep, 2019
具对抗性破坏的随机线性优化
Stochastic Linear Optimization with Adversarial Corruption
HTML
PDF
Yingkai Li, Edmund Y. Lou, Liren Shan
TL;DR
将随机赌徒模型扩展到对抗性破坏模型，并应用于随机线性优化问题，采用L"owner-John椭球进行探索，并将时间分割成指数递增的时代以限制破坏的影响，使得算法对于适应性对手所选择的破坏的数量不可知，其遗憾只会在破坏数量线性增加时增加。
Abstract
We extend the model of
stochastic bandits
with
adversarial corruption
(Lykouriset al., 2018) to the
stochastic linear optimization
problem
→