arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2303.03327cs.LGstat.ML
收藏

基于决策-估计系数的$\gamma$-悔值紧界

Tight Bounds for $γ$-Regret via the Decision-Estimation Coefficient

Margalit Glasgow, Alexander Rakhlin

AI总结:

本研究针对结构化老虎机中精确最优解难求的问题,提出修正的统计复杂度参数$\gamma$-DEC,用新技术推导了$\gamma$-悔值的匹配上下界,明确了该悔值的基本极限。

摘要:

在本研究中,我们对任意结构化老虎机问题的$\gamma$-悔值给出了统计刻画,$\gamma$-悔值是指与最优解$\gamma$倍大小的基准比较时产生的悔值。$\gamma$-悔值出现在函数类$\mathcal{F}$上的结构化老虎机问题中,这类问题里找到$f \in \mathcal{F}$的精确最优解是难以处理的。我们的刻画通过$\gamma$-DEC(即函数类$\mathcal{F}$的统计复杂度参数)给出,它是Foster等人2023年提出的约束决策-估计系数(DEC)的修正版本,也与Foster等人2021年提出的原始偏移DEC密切相关。我们的下界表明,$\gamma$-DEC是任意模型类$\mathcal{F}$的基本极限:对于任何算法,都存在某个$f \in \mathcal{F}$,使得该算法的$\gamma$-悔值(几乎)与$\mathcal{F}$的$\gamma$-DEC成比例。我们还给出了上界,证明存在一种算法能达到几乎匹配的$\gamma$-悔值。由于将DEC的已有结果应用到$\gamma$-悔值场景存在重大挑战,我们的上下界推导都需要全新的技术和一种新算法。

英文摘要:

In this work, we give a statistical characterization of the $γ$-regret for arbitrary structured bandit problems, the regret which arises when comparing against a benchmark that is $γ$ times the optimal solution. The $γ$-regret emerges in structured bandit problems over a function class $\mathcal{F}$ where finding an exact optimum of $f \in \mathcal{F}$ is intractable. Our characterization is given in terms of the $γ$-DEC, a statistical complexity parameter for the class $\mathcal{F}$, which is a modification of the constrained Decision-Estimation Coefficient (DEC) of Foster et al., 2023 (and closely related to the original offset DEC of Foster et al., 2021). Our lower bound shows that the $γ$-DEC is a fundamental limit for any model class $\mathcal{F}$: for any algorithm, there exists some $f \in \mathcal{F}$ for which the $γ$-regret of that algorithm scales (nearly) with the $γ$-DEC of $\mathcal{F}$. We provide an upper bound showing that there exists an algorithm attaining a nearly matching $γ$-regret. Due to significant challenges in applying the prior results on the DEC to the $γ$-regret case, both our lower and upper bounds require novel techniques and a new algorithm.

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。

↑