arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2102.01868cs.IRcs.AIcs.LG
收藏

因果协同过滤

Causal Collaborative Filtering

Rutgers University(罗格斯大学);Renmin University of China(中国人民大学)

Shuyuan Xu, Yingqiang Ge, Yunqi Li, Zuohui Fu, Xu Chen, Yongfeng Zhang

AI总结:

针对协同过滤中纯相关学习引发的辛普森悖论,提出 CCF 因果框架,通过条件干预与反事实约束学习估计因果偏好,并在真实及随机试验数据上验证其性能提升。

摘要:

许多传统推荐算法的设计基于一个基本思想,即从数据中挖掘或学习相关模式,以估计用户—物品的相关偏好。然而,纯粹的相关学习可能在预测中导致辛普森悖论,从而牺牲推荐性能。辛普森悖论是一种众所周知的统计现象,会造成统计结论上的混淆;忽视该悖论可能导致不准确的决策。幸运的是,因果与反事实建模能够帮助我们跳出观测数据进行用户建模和个性化,从而解决此类问题。本文提出 Causal Collaborative Filtering(CCF,因果协同过滤),这是一个在协同过滤与推荐中建模因果性的通用框架。我们给出了协同过滤(CF)的统一因果视角,并从数学上表明,许多传统 CF 算法实际上是 CCF 在简化因果图下的特例。随后,我们提出一种针对 $do$-运算的条件干预方法,使我们能够基于观测数据估计用户—物品的因果偏好。最后,我们进一步提出一个用于估计用户—物品偏好的通用反事实约束学习框架。实验在两类真实世界数据集上进行——传统数据集和随机试验数据集——结果表明,我们的框架能够提升推荐性能,并减轻许多 CF 算法中的辛普森悖论问题。

英文摘要:

Many of the traditional recommendation algorithms are designed based on the fundamental idea of mining or learning correlative patterns from data to estimate the user-item correlative preference. However, pure correlative learning may lead to Simpson's paradox in predictions, and thus results in sacrificed recommendation performance. Simpson's paradox is a well-known statistical phenomenon, which causes confusions in statistical conclusions and ignoring the paradox may result in inaccurate decisions. Fortunately, causal and counterfactual modeling can help us to think outside of the observational data for user modeling and personalization so as to tackle such issues. In this paper, we propose Causal Collaborative Filtering (CCF) -- a general framework for modeling causality in collaborative filtering and recommendation. We provide a unified causal view of CF and mathematically show that many of the traditional CF algorithms are actually special cases of CCF under simplified causal graphs. We then propose a conditional intervention approach for $do$-operations so that we can estimate the user-item causal preference based on the observational data. Finally, we further propose a general counterfactual constrained learning framework for estimating the user-item preferences. Experiments are conducted on two types of real-world datasets -- traditional and randomized trial data -- and results show that our framework can improve the recommendation performance and reduce the Simpson's paradox problem of many CF algorithms.

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。

↑