AdaF^2M^2: Comprehensive Learning and Responsive Leveraging Features in Recommendation System

TL;DR

This paper introduces AdaF^2M^2, a model-agnostic framework that improves feature learning and leveraging in recommendation systems by addressing long-tail data distribution issues, resulting in better generalization and performance.

Contribution

The paper proposes AdaF^2M^2, a novel adaptive feature modeling framework with feature masking and weighting, enhancing feature representation and leveraging in recommendation systems.

Findings

Online A/B tests show +1.37% and +1.89% improvements in user active days and app duration.

Offline experiments demonstrate consistent performance gains across models.

Widely deployed in Douyin, confirming effectiveness and universality.

Abstract

Feature modeling, which involves feature representation learning and leveraging, plays an essential role in industrial recommendation systems. However, the data distribution in real-world applications usually follows a highly skewed long-tail pattern due to the popularity bias, which easily leads to over-reliance on ID-based features, such as user/item IDs and ID sequences of interactions. Such over-reliance makes it hard for models to learn features comprehensively, especially for those non-ID meta features, e.g., user/item characteristics. Further, it limits the feature leveraging ability in models, getting less generalized and more susceptible to data noise. Previous studies on feature modeling focus on feature extraction and interaction, hardly noticing the problems brought about by the long-tail data distribution. To achieve better feature representation learning and leveraging on real-world data, we propose a model-agnostic framework AdaF^2M^2, short for Adaptive Feature Modeling with Feature Mask. The feature-mask mechanism helps comprehensive feature learning via multi-forward training with augmented samples, while the adapter applies adaptive weights on features responsive to different user/item states. By arming base models with AdaF^2M^2, we conduct online A/B tests on multiple recommendation scenarios, obtaining +1.37% and +1.89% cumulative improvements on user active days and app duration respectively. Besides, the extended offline experiments on different models show improvements as well. AdaF$^2$M$^2$ has been widely deployed on both retrieval and ranking tasks in multiple applications of Douyin Group, indicating its superior effectiveness and universality.