Towards Automated Neural Interaction Discovery for Click-Through Rate Prediction

TL;DR

This paper introduces AutoCTR, an automated framework using neural architecture search to discover effective interaction architectures for CTR prediction, addressing challenges of feature diversity and data volume.

Contribution

AutoCTR automates the discovery of neural interaction architectures for CTR prediction, combining evolutionary search with learning-to-rank guidance and low-fidelity acceleration.

Findings

AutoCTR outperforms human-crafted architectures on multiple datasets.

Discovered architectures show strong generalizability and transferability.

AutoCTR accelerates architecture search with low-fidelity models.

Abstract

Click-Through Rate (CTR) prediction is one of the most important machine learning tasks in recommender systems, driving personalized experience for billions of consumers. Neural architecture search (NAS), as an emerging field, has demonstrated its capabilities in discovering powerful neural network architectures, which motivates us to explore its potential for CTR predictions. Due to 1) diverse unstructured feature interactions, 2) heterogeneous feature space, and 3) high data volume and intrinsic data randomness, it is challenging to construct, search, and compare different architectures effectively for recommendation models. To address these challenges, we propose an automated interaction architecture discovering framework for CTR prediction named AutoCTR. Via modularizing simple yet representative interactions as virtual building blocks and wiring them into a space of direct acyclic graphs, AutoCTR performs evolutionary architecture exploration with learning-to-rank guidance at the architecture level and achieves acceleration using low-fidelity model. Empirical analysis demonstrates the effectiveness of AutoCTR on different datasets comparing to human-crafted architectures. The discovered architecture also enjoys generalizability and transferability among different datasets.