In recommender systems, various latent confounding factors (e.g., user social environment and item public attractiveness) can affect user behavior, item exposure, and feedback in distinct ways. These factors may directly or indirectly impact user feedback and are often shared across items or users, making them multi-cause latent confounders. However, existing methods typically fail to account for latent confounders between users and their feedback, as well as those between items and user feedback simultaneously. To address the problem of multi-cause latent confounders, we propose a multi-cause deconfounding method for recommender systems with latent confounders (MCDCF). MCDCF leverages multi-cause causal effect estimation to learn substitutes for latent confounders associated with both users and items, using user behaviour data. Specifically, MCDCF treats the multiple items that users interact with and the multiple users that interact with items as treatment variables, enabling it to learn substitutes for the latent confounders that influence the estimation of causality between users and their feedback, as well as between items and user feedback. Additionally, we theoretically demonstrate the soundness of our MCDCF method. Extensive experiments on three real-world datasets demonstrate that our MCDCF method effectively recovers latent confounders related to users and items, reducing bias and thereby improving recommendation accuracy.
The Deconfounded Recommender: A Causal Inference Approach to Recommendation
Columbia University（哥伦比亚大学）；Netflix Inc.（奈飞公司）；Mila；HEC Montréal（蒙特利尔高等商学院）
推荐的目标是向用户展示他们喜欢的项目。虽然通常被框定为预测，但推荐的精神是回答一个介入性的问题-对于每个用户和电影，如果我们“强迫”用户观看电影，评分会是多少？为此，我们开发了一种因果推荐方法，其中观看电影是一种“治疗”，用户的评级是一种“结果”。“问题是可能存在未观察到的混杂因素，这些变量影响用户观看的电影以及他们如何评价它们;未观察到的混杂因素阻碍了观察数据的因果预测。为了解决这个问题，我们开发了去基推荐器，一种使用经典推荐模型进行因果推荐的方法。在Wang & Blei [23]之后，去基推荐器涉及两个概率模型。第一个模型是用户观看的电影;它为未观察到的混杂因素提供了替代品。第二个模型模拟每个用户如何评价每部电影;它使用替代品来帮助解释混淆因素。这种两阶段方法消除了由于混淆而导致的偏倚。它改善了推荐，并在测试集上具有稳定的性能。
Causal Structure Representation Learning of Confounders in Latent Space for Recommendation
MIC Lab, College of Computer Science and Technology, Jilin University（吉林大学计算机科学与技术学院MIC实验室）；Key Laboratory of Trustworthy Distributed Computing and Service (MoE), Beijing University of Posts and Telecommunications（北京邮电大学可信分布式计算与服务重点实验室）；Institute of Artificial Intelligence, Beihang University and Zhongguancun Laboratory（北航人工智能研究院和中关村实验室）
Probabilistic Residual Learning for Online Recommendations
Rutgers University Piscataway New Jersey United States（Meta）；Meta Sunnyvale California United States；University of Copenhagen Copenhagen Denmark；Rutgers University \& UIUC Piscataway New Jersey United States；Rutgers University；Meta；University of Copenhagen；Rutgers University \& UIUC
National University of Singapore Master of Industrial and Systems Engineering（新加坡国立大学工业与系统工程硕士）；East China Normal University Master of Library and Information Science（华东师范大学图书馆与信息科学硕士）；Shandong Normal University（山东师范大学）
