Ad-Rec: Advanced Feature Interactions to Address Covariate-Shifts in Recommendation Networks

TL;DR

Ad-Rec introduces a transformer-based recommendation network that effectively learns higher-order feature interactions to combat covariate shifts, improving model performance and training efficiency.

Contribution

This paper presents Ad-Rec, a novel transformer-based model that enhances feature interaction learning to address covariate shifts in recommendation systems.

Findings

Improves AUC performance on recommendation tasks

Accelerates model convergence and reduces training time

Demonstrates scalability and effectiveness through ablation studies

Abstract

Recommendation models are vital in delivering personalized user experiences by leveraging the correlation between multiple input features. However, deep learning-based recommendation models often face challenges due to evolving user behaviour and item features, leading to covariate shifts. Effective cross-feature learning is crucial to handle data distribution drift and adapting to changing user behaviour. Traditional feature interaction techniques have limitations in achieving optimal performance in this context. This work introduces Ad-Rec, an advanced network that leverages feature interaction techniques to address covariate shifts. This helps eliminate irrelevant interactions in recommendation tasks. Ad-Rec leverages masked transformers to enable the learning of higher-order cross-features while mitigating the impact of data distribution drift. Our approach improves model quality, accelerates convergence, and reduces training time, as measured by the Area Under Curve (AUC) metric. We demonstrate the scalability of Ad-Rec and its ability to achieve superior model quality through comprehensive ablation studies.