Propensity-Independent Bias Recovery in Offline Learning-to-Rank Systems
Propensity-Independent Bias Recovery in Offline Learning-to-Rank Systems
复制标题
DOI：
10.1145/3404835.3463097
复制
发表时间：
2021-07
期刊：
影响因子：
--
通讯作者：
Zohreh Ovaisi;K. Vasilaky;E. Zheleva
中科院分区：
文献类型：
--
作者：
Zohreh Ovaisi;K. Vasilaky;E. Zheleva
学习排名系统通常利用用户-项目交互数据（例如，点击），为用户提供高质量的排名。然而，这些数据存在一些偏差，如果天真地用作训练数据，它可能会导致次优排名算法。大多数现有的偏差校正方法都集中在位置偏差上，即排名较高的结果更有可能接受交互，并通过利用反向倾向加权来解决这种偏差。然而，它并不总是能够准确地估计倾向分数，除了位置偏差，选择偏差经常遇到在现实世界中的推荐系统。选择偏差的发生是因为用户暴露在一个截断的结果列表中，这使得一些项目被观察到的机会为零，因此，即使它们是相关的，也没有机会进行交互。在这里，我们提出了一个新的反事实的方法，使用两个阶段的校正方法，并共同解决选择和位置偏差的学习排名系统，而不依赖于倾向分数。我们的实验结果表明，我们的方法优于最先进的倾向独立的方法，或者优于或可比的方法，使强假设的倾向模型是已知的。
Learning-to-rank systems often utilize user-item interaction data (e.g., clicks) to provide users with high-quality rankings. However, this data suffers from several biases, and if naively used as training data, it can lead to suboptimal ranking algorithms. Most existing bias-correcting methods focus on position bias, the fact that higher-ranked results are more likely to receive interaction, and address this bias by leveraging inverse propensity weighting. However, it is not always possible to accurately estimate propensity scores, and in addition to position bias, selection bias is often encountered in real-world recommender systems. Selection bias occurs because users are exposed to a truncated list of results, which gives a zero chance for some items to be observed and, therefore, interacted with, even if they are relevant. Here, we propose a new counterfactual method that uses a two-stage correction approach and jointly addresses selection and position bias in learning-to-rank systems without relying on propensity scores. Our experimental results show that our method is better than state-of-the-art propensity-independent methods and either better than or comparable to methods that make the strong assumption for which the propensity model is known.