Star+: A New Multi-Domain Model for CTR Prediction

TL;DR

Star+ is a multi-domain CTR prediction model that enhances domain-specific and shared information interaction using various fusion and normalization techniques, leading to improved accuracy and efficiency.

Contribution

It introduces a novel multi-domain model with fusion strategies and normalization techniques, advancing multi-domain CTR prediction methods.

Findings

Star+ outperforms existing models on industrial and public datasets.

Fusion strategies improve the balance between shared and domain-specific data.

Normalization techniques impact model performance significantly.

Abstract

In this paper, we introduce Star+, a novel multi-domain model for click-through rate (CTR) prediction inspired by the Star model. Traditional single-domain approaches and existing multi-task learning techniques face challenges in multi-domain environments due to their inability to capture domain-specific data distributions and complex inter-domain relationships. Star+ addresses these limitations by enhancing the interaction between shared and domain-specific information through various fusion strategies, such as add, adaptive add, concatenation, and gating fusions, to find the optimal balance between domain-specific and shared information. We also investigate the impact of different normalization techniques, including layer normalization, batch normalization, and partition normalization, on the performance of our model. Our extensive experiments on both industrial and public datasets demonstrate that Star+ significantly improves prediction accuracy and efficiency. This work contributes to the advancement of recommendation systems by providing a robust, scalable, and adaptive solution for multi-domain environments.