Ad-Rec: Advanced Feature Interactions to Address Covariate-Shifts in Recommendation Networks
This addresses the problem of evolving user behavior and item features causing data distribution drift in recommendation systems, though it appears incremental as it builds on existing feature interaction techniques.
The paper tackles covariate shifts in deep learning-based recommendation models by introducing Ad-Rec, which uses masked transformers for advanced feature interactions, resulting in improved model quality, faster convergence, and reduced training time as measured by AUC.
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.