AIM: Automatic Interaction Machine for Click-Through Rate Prediction

TL;DR

AIM is a unified framework that automatically searches for significant feature interactions, appropriate interaction functions, and embedding dimensions, significantly improving CTR prediction performance in large-scale online systems.

Contribution

The paper introduces AIM, a novel method that automates feature interaction search, interaction function selection, and embedding dimension search for CTR prediction.

Findings

AIM outperforms baseline models on three large-scale datasets.

AIM improves DeepFM CTR by 4.4% in online A/B testing.

The framework effectively prunes useless feature interactions.

Abstract

Feature embedding learning and feature interaction modeling are two crucial components of deep models for Click-Through Rate (CTR) prediction. Most existing deep CTR models suffer from the following three problems. First, feature interactions are either manually designed or simply enumerated. Second, all the feature interactions are modeled with an identical interaction function. Third, in most existing models, different features share the same embedding size which leads to memory inefficiency. To address these three issues mentioned above, we propose Automatic Interaction Machine (AIM) with three core components, namely, Feature Interaction Search (FIS), Interaction Function Search (IFS) and Embedding Dimension Search (EDS), to select significant feature interactions, appropriate interaction functions and necessary embedding dimensions automatically in a unified framework. Specifically, FIS component automatically identifies different orders of essential feature interactions with useless ones pruned; IFS component selects appropriate interaction functions for each individual feature interaction in a learnable way; EDS component automatically searches proper embedding size for each feature. Offline experiments on three large-scale datasets validate the superior performance of AIM. A three-week online A/B test in the recommendation service of a mainstream app market shows that AIM improves DeepFM model by 4.4% in terms of CTR.