AutoFAS: Automatic Feature and Architecture Selection for Pre-Ranking System

TL;DR

AutoFAS introduces a joint neural architecture and feature selection framework for pre-ranking systems, optimizing efficiency and effectiveness simultaneously, leading to improved performance with lower computational costs in industrial search applications.

Contribution

AutoFAS is the first to jointly optimize feature selection and network architecture using NAS guided by ranking performance, addressing efficiency-effectiveness trade-offs.

Findings

Outperforms previous SOTA methods in real-world search system.

Reduces computational cost while improving pre-ranking accuracy.

Successfully deployed in Meituan's search system with significant gains.

Abstract

Industrial search and recommendation systems mostly follow the classic multi-stage information retrieval paradigm: matching, pre-ranking, ranking, and re-ranking stages. To account for system efficiency, simple vector-product based models are commonly deployed in the pre-ranking stage. Recent works consider distilling the high knowledge of large ranking models to small pre-ranking models for better effectiveness. However, two major challenges in pre-ranking system still exist: (i) without explicitly modeling the performance gain versus computation cost, the predefined latency constraint in the pre-ranking stage inevitably leads to suboptimal solutions; (ii) transferring the ranking teacher's knowledge to a pre-ranking student with a predetermined handcrafted architecture still suffers from the loss of model performance. In this work, a novel framework AutoFAS is proposed which jointly optimizes the efficiency and effectiveness of the pre-ranking model: (i) AutoFAS for the first time simultaneously selects the most valuable features and network architectures using Neural Architecture Search (NAS) technique; (ii) equipped with ranking model guided reward during NAS procedure, AutoFAS can select the best pre-ranking architecture for a given ranking teacher without any computation overhead. Experimental results in our real world search system show AutoFAS consistently outperforms the previous state-of-the-art (SOTA) approaches at a lower computing cost. Notably, our model has been adopted in the pre-ranking module in the search system of Meituan, bringing significant improvements.