arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1902.08593cs.LGstat.APstat.ML
收藏

非平稳报酬分布和时滞反馈过程的多臂Bandit策略

Multi-Armed Bandit Strategies for Non-Stationary Reward Distributions and Delayed Feedback Processes

Loblaw Digital(罗布劳数字公司)

Larkin Liu, Richard Downe, Joshua Reid

摘要:

进行调查的各种多臂强盗(MAB)的策略,以检查他们的表现,表现出非平稳的随机奖励函数的情况下,与延迟反馈。我们运行了几个MAB模拟来模拟杂货店提货的在线电子商务平台,优化产品可用性。在这项工作中,我们评估了几个流行的MAB策略,如$ε$ -贪婪,UCB 1,和汤普森采样。我们比较各自的表现,每一个MAB策略的背景下,后悔最小化。我们在奖励函数是非平稳的情况下进行分析。此外,该过程经历延迟反馈,其中奖励功能不会立即响应所玩的手臂。我们设计了一个新的自适应技术(AG 1)量身定制的非平稳的奖励函数在延迟反馈的情况下。仿真结果表明,与传统的MAB策略相比,该策略在遗憾最小化方面具有更好的性能。

英文摘要:

A survey is performed of various Multi-Armed Bandit (MAB) strategies in order to examine their performance in circumstances exhibiting non-stationary stochastic reward functions in conjunction with delayed feedback. We run several MAB simulations to simulate an online eCommerce platform for grocery pick up, optimizing for product availability. In this work, we evaluate several popular MAB strategies, such as $ε$-greedy, UCB1, and Thompson Sampling. We compare the respective performances of each MAB strategy in the context of regret minimization. We run the analysis in the scenario where the reward function is non-stationary. Furthermore, the process experiences delayed feedback, where the reward function is not immediately responsive to the arm played. We devise a new adaptive technique (AG1) tailored for non-stationary reward functions in the delayed feedback scenario. The results of the simulation show show superior performance in the context of regret minimization compared to traditional MAB strategies.

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。

↑