RE-SORT: Removing Spurious Correlation in Multilevel Interaction for CTR Prediction

Songli Wu, Liang Du, Jia-Qi Yang, Yuai Wang, De-Chuan Zhan, SHUANG ZHAO, Zixun Sun

Keywords: click through rate, recommendation system, spurious correlation, stacked recurrent network

TL;DR: We propose a CTR prediction framework that removes spurious correlations in multilevel feature interactions, which leverages critical causal relationships between items and users in diverse nonlinear feature spaces to enhance the CTR prediction.

Abstract: Click-through rate (CTR) prediction is a critical task in recommendation systems, serving as the ultimate filtering step to sort items for a user. Most recent cutting-edge methods primarily focus on investigating complex implicit and explicit feature interactions; however, these methods neglect the spurious correlation issue caused by confounding factors, thereby diminishing the model's generalization ability. We propose a CTR prediction framework that REmoves Spurious cORrelations in mulTilevel feature interactions, termed RE-SORT, which has two key components. I. A multilevel stacked recurrent (MSR) structure enables the model to efficiently capture diverse nonlinear interactions from feature spaces at different levels. II. A spurious correlation elimination (SCE) module further leverages Laplacian kernel mapping and sample reweighting methods to eliminate the spurious correlations concealed within the multilevel features, allowing the model to focus on the true causal features. Extensive experiments conducted on four challenging CTR datasets, our production dataset, and an online A/B test demonstrate that the proposed method achieves state-of-the-art performance in both accuracy and speed. The utilized codes, models, and dataset will be released at https://github.com/RE-SORT.

List Of Authors: Wu, Songli and Du, Liang and Yang, Jiaqi and Wang, Yuai and Zhan, Dechuan and Zhao, Shuang and Sun, Zixun

Submission Number: 11