arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2008.07146cs.LGstat.ML
收藏

开放Bandit数据集和管道:实现现实和可复制的非政策评估

Open Bandit Dataset and Pipeline: Towards Realistic and Reproducible Off-Policy Evaluation

Hanjuku-kaso Co Ltd(Hanjuku-kaso有限公司);Cornell University(康奈尔大学);ZOZO Research(ZOZO研究院);Yale University(耶鲁大学)

Yuta Saito, Shunsuke Aihara, Megumi Matsutani, Yusuke Narita

摘要:

政策外评估(OPE)的目的是使用不同政策产生的数据来估计假设政策的性能。由于其在实践中的巨大潜在影响,人们对这一领域的研究兴趣越来越大。然而,没有真实世界的公共数据集可以评估OPE,这使得其实验研究不现实且不可复制。为了实现现实和可重复的OPE研究,我们提出了开放的Bandit数据集,这是一个在大型时尚电子商务平台ZOZOTOWN上收集的公共记录的Bandit数据集。我们的数据集是独一无二的,因为它包含一组通过在同一平台上运行不同策略收集的多个记录的bandit数据集。这使得实验比较不同的OPE估计的第一次。我们还开发了名为Open Bandit Pipeline的Python软件,以简化和标准化批处理Bandit算法和OPE的实现。我们的开放数据和软件将有助于公平和透明的OPE研究,并帮助社区确定富有成效的研究方向。我们使用我们的数据集和软件提供了现有OPE估计器的广泛基准实验。这些结果为未来的OPE研究开辟了必要的挑战和新的途径。

英文摘要:

Off-policy evaluation (OPE) aims to estimate the performance of hypothetical policies using data generated by a different policy. Because of its huge potential impact in practice, there has been growing research interest in this field. There is, however, no real-world public dataset that enables the evaluation of OPE, making its experimental studies unrealistic and irreproducible. With the goal of enabling realistic and reproducible OPE research, we present Open Bandit Dataset, a public logged bandit dataset collected on a large-scale fashion e-commerce platform, ZOZOTOWN. Our dataset is unique in that it contains a set of multiple logged bandit datasets collected by running different policies on the same platform. This enables experimental comparisons of different OPE estimators for the first time. We also develop Python software called Open Bandit Pipeline to streamline and standardize the implementation of batch bandit algorithms and OPE. Our open data and software will contribute to fair and transparent OPE research and help the community identify fruitful research directions. We provide extensive benchmark experiments of existing OPE estimators using our dataset and software. The results open up essential challenges and new avenues for future OPE research.

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。

↑