arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2007.08584stat.MLcs.LG
收藏

未知协方差下的自校正Bandits

Self-Tuning Bandits over Unknown Covariate-Shifts

Columbia University(哥伦比亚大学)

Joseph Suk, Samory Kpotufe

摘要:

有协变量的强盗,也就是。上下文强盗(contextual bandits)解决了在给定时间$t$的最佳动作(或武装)取决于上下文$x_t$的情况,例如,一个新病人的病史一个消费者过去的购买记录虽然可以理解上下文的分布可能随时间而改变,由于季节性,或部署到新的环境中,大量的研究关注的是最不利的这种变化,导致遗憾的界限,往往是最坏的情况下的性质。另一方面,协变量偏移在分类中被认为是一种中间形式,可以捕获分布中的轻微到相对严重的变化。我们认为非参数土匪在这样的中间地带的情况下,并得出新的遗憾的界限,紧紧抓住上下文分布的连续变化。此外,我们表明,这些速率可以自适应地达到没有知识的时间的转变,也没有转移量。

英文摘要:

Bandits with covariates, a.k.a. contextual bandits, address situations where optimal actions (or arms) at a given time $t$, depend on a context $x_t$, e.g., a new patient's medical history, a consumer's past purchases. While it is understood that the distribution of contexts might change over time, e.g., due to seasonalities, or deployment to new environments, the bulk of studies concern the most adversarial such changes, resulting in regret bounds that are often worst-case in nature. Covariate-shift on the other hand has been considered in classification as a middle-ground formalism that can capture mild to relatively severe changes in distributions. We consider nonparametric bandits under such middle-ground scenarios, and derive new regret bounds that tightly capture a continuum of changes in context distribution. Furthermore, we show that these rates can be adaptively attained without knowledge of the time of shift nor the amount of shift.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑