AutoField: Automating Feature Selection in Deep Recommender Systems

TL;DR

AutoField introduces an automated feature selection framework for deep recommender systems, reducing manual effort and improving recommendation efficiency by adaptively choosing essential features using a differentiable controller network.

Contribution

The paper presents a novel AutoML framework that automatically selects important feature fields in deep recommender systems, enhancing efficiency and reducing manual feature engineering.

Findings

Effective feature selection improves recommendation performance.

Framework demonstrates transferability across datasets.

Key components and parameters influence selection accuracy.

Abstract

Feature quality has an impactful effect on recommendation performance. Thereby, feature selection is a critical process in developing deep learning-based recommender systems. Most existing deep recommender systems, however, focus on designing sophisticated neural networks, while neglecting the feature selection process. Typically, they just feed all possible features into their proposed deep architectures, or select important features manually by human experts. The former leads to non-trivial embedding parameters and extra inference time, while the latter requires plenty of expert knowledge and human labor effort. In this work, we propose an AutoML framework that can adaptively select the essential feature fields in an automatic manner. Specifically, we first design a differentiable controller network, which is capable of automatically adjusting the probability of selecting a particular feature field; then, only selected feature fields are utilized to retrain the deep recommendation model. Extensive experiments on three benchmark datasets demonstrate the effectiveness of our framework. We conduct further experiments to investigate its properties, including the transferability, key components, and parameter sensitivity.