English(EN)Sequential Batch Learning in Finite-Action Linear Contextual Bandits
新研究探讨老虎机算法以改进决策和减少遗憾 · 跟踪 8 个来源
作者PulseAugur 编辑部·[8 个来源]·
几篇近期研究论文探讨了老虎机算法的进展，这是一种顺序决策框架。一篇论文介绍了潜在顺序老虎机（LOB），它通过仅要求了解状态内动作偏好的部分顺序来放宽先前潜在老虎机算法的假设，从而提高样本效率。另一项研究侧重于多臂老虎机中遗憾与不稳定性之间的权衡，提出了一种新的算法 SLE-UCB，该算法匹配理论下界。进一步的研究解决了具有任意反馈延迟的 Lipschitz 老虎机，开发了实现强遗憾保证的算法，并探讨了容量受限延迟老虎机优化中的条件能量和时间几何，揭示了基于时间和容量的遗憾的细微差别。最后，一篇关于上下文老虎机的论文提出了一种快速、同类最佳的遗憾算法，另一篇则研究了线性上下文老虎机中的顺序批量学习，为实际应用提供了近乎完整的表征。
AI
arXiv:2605.07304v2 Announce Type: replace Abstract: Bandit algorithms solve diverse sequential decision-making problems, but are often too sample-inefficient for from-scratch personalization. To substantially reduce exploration times, latent bandit algorithms exploit cross-instan…
Multi-armed bandit algorithms are evaluated by regret, yet comparable regret can coexist with different allocations across independent runs. We study the trade-off between worst-case regret $\mathcal{R}_{K,T}$ and instability $\mathcal S_{K,T}$, defined as the largest standard de…
arXiv:2608.15036v1 Announce Type: new Abstract: The Lipschitz bandit problem extends the traditional multi-armed bandit framework to continuous action spaces by assuming that the reward functions satisfy a Lipschitz condition. This work investigates Lipschitz bandits under arbitr…
arXiv cs.LG
TIER_1English(EN)·Anling Xiang, Yuwen Yang, Yang Shen·
arXiv:2608.16216v1 Announce Type: new Abstract: What is the right delay complexity when a learner can track only $C$ pending feedback items and discarded feedback is permanently lost? Existing one-point bandit convex optimization guarantees in this model pay $\sqrt{T\sigma_{\max}…
arXiv:2510.15483v3 Announce Type: replace Abstract: We study the problem of stochastic contextual bandits in the agnostic setting, where the goal is to compete with the best policy in a given class without assuming realizability or imposing model restrictions on losses or rewards…
arXiv stat.ML
TIER_1English(EN)·Kaifei Wang, Yinyu Ye, Han Zhong·
arXiv:2608.17841v1 Announce Type: new Abstract: Multi-armed bandit algorithms are evaluated by regret, yet comparable regret can coexist with different allocations across independent runs. We study the trade-off between worst-case regret $\mathcal{R}_{K,T}$ and instability $\math…
arXiv:2608.15848v1 Announce Type: new Abstract: We study generalized linear bandits with memory, an endogenous non-stationary setting in which rewards depend on past actions through a finite memory matrix. Building on prior work for linear models (Clerici et al., 2024), we show t…
arXiv stat.ML
TIER_1English(EN)·Yanjun Han, Zhengqing Zhou, Zihao Hu, Jose Blanchet, Peter W. Glynn, Yinyu Ye, Zhengyuan Zhou·
arXiv:2004.06321v2 Announce Type: replace-cross Abstract: We study the sequential batch learning problem in linear contextual bandits with finite action sets, where the decision maker is constrained to split incoming individuals into (at most) a fixed number of batches and can on…
