Blockwise Feature Interaction in Recommendation Systems

TL;DR

This paper introduces blockwise feature interaction (BFI), a method that reduces computational costs in recommendation systems by partitioning feature interactions, maintaining accuracy while improving efficiency.

Contribution

It proposes a novel blockwise approach with four variants that significantly decreases memory and computation in feature interaction models without sacrificing accuracy.

Findings

BFI variants achieve comparable accuracy to DCNv2

Substantial reduction in computational overhead

Fewer parameters needed for similar performance

Abstract

Feature interactions can play a crucial role in recommendation systems as they capture complex relationships between user preferences and item characteristics. Existing methods such as Deep & Cross Network (DCNv2) may suffer from high computational requirements due to their cross-layer operations. In this paper, we propose a novel approach called blockwise feature interaction (BFI) to help alleviate this issue. By partitioning the feature interaction process into smaller blocks, we can significantly reduce both the memory footprint and the computational burden. Four variants (denoted by P, Q, T, S, respectively) of BFI have been developed and empirically compared. Our experimental results demonstrate that the proposed algorithms achieves close accuracy compared to the standard DCNv2, while greatly reducing the computational overhead and the number of parameters. This paper contributes to the development of efficient recommendation systems by providing a practical solution for improving feature interaction efficiency.