Open Bandit Dataset and Pipeline: Towards Realistic and Reproducible Off-Policy Evaluation
This work addresses a critical bottleneck for researchers in reinforcement learning and recommendation systems by providing standardized tools for fair and transparent OPE research.
The authors tackled the lack of real-world public datasets for off-policy evaluation (OPE) by introducing the Open Bandit Dataset, collected from a fashion e-commerce platform, and the Open Bandit Pipeline software, enabling realistic and reproducible experiments for the first time.
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.