← 研踪 PaperTrailarXiv:2611.11494 · v1
Regret without Optimal-Arm Feedback under Reward-Dependent Delays
首次提交 2026-11-09 00:44
版本更新 2026-11-09 00:44
雷达观测 2026-11-10 12:16
摘要
We study stochastic bandits with reward-dependent delays, possibly infinite, whose reports omit the originating round. For each $c\in[0,1)$, Balanced Historical Optimism (BHO) has a delay term controlled by the largest suboptimal-arm quantile at level $1-cΔ_i$, without requiring optimal-arm reports. Lower bounds give an inverse-square slack cost in the logarithmic statistical regime and linear minimax regret at $c=1$. A family with growing $K$ forces worst-case regret $Ω(G\log K)$ for every learner, where $G=\max_iΔ_i h_i$ is the largest gap-weighted suboptimal cutoff. Pending-triggered balance preserves BHO's weighted-cutoff bound for $c<1$ and attains the historical-greedy exploration coefficient with deterministic rewards, a unique optimum and fixed bounded delays. Gap-weighted comparisons and experiments show when shared waiting helps and what balanced exploration costs.
作者备注：45 pages, 3 figures. Code and data are included as ancillary files
观测到的版本
相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-09 00:44 · 最新观测版本
Regret without Optimal-Arm Feedback under Reward-Dependent Delays
We study stochastic bandits with reward-dependent delays, possibly infinite, whose reports omit the originating round. For each $c\in[0,1)$, Balanced Historical Optimism (BHO) has a delay term controlled by the largest suboptimal-arm quantile at level $1-cΔ_i$, without requiring optimal-arm reports. Lower bounds give an inverse-square slack cost in the logarithmic statistical regime and linear minimax regret at $c=1$. A family with growing $K$ forces worst-case regret $Ω(G\log K)$ for every learner, where $G=\max_iΔ_i h_i$ is the largest gap-weighted suboptimal cutoff. Pending-triggered balance preserves BHO's weighted-cutoff bound for $c<1$ and attains the historical-greedy exploration coefficient with deterministic rewards, a unique optimum and fixed bounded delays. Gap-weighted comparisons and experiments show when shared waiting helps and what balanced exploration costs.
查看此版本解读 →查看此版本原文 ↗