EulerNet: Adaptive Feature Interaction Learning via Euler's Formula for CTR Prediction

TL;DR

EulerNet introduces an adaptive method for high-order feature interaction learning in CTR prediction, leveraging Euler's formula to efficiently model complex interactions without predefining interaction orders, resulting in improved effectiveness and efficiency.

Contribution

The paper proposes EulerNet, a novel model that adaptively learns high-order feature interactions in a complex space using Euler's formula, eliminating the need for manual interaction order design.

Findings

EulerNet outperforms existing methods on three public datasets.

The approach significantly reduces computational costs.

EulerNet enhances model capability by combining implicit and explicit interactions.

Abstract

Learning effective high-order feature interactions is very crucial in the CTR prediction task. However, it is very time-consuming to calculate high-order feature interactions with massive features in online e-commerce platforms. Most existing methods manually design a maximal order and further filter out the useless interactions from them. Although they reduce the high computational costs caused by the exponential growth of high-order feature combinations, they still suffer from the degradation of model capability due to the suboptimal learning of the restricted feature orders. The solution to maintain the model capability and meanwhile keep it efficient is a technical challenge, which has not been adequately addressed. To address this issue, we propose an adaptive feature interaction learning model, named as EulerNet, in which the feature interactions are learned in a complex vector space by conducting space mapping according to Euler's formula. EulerNet converts the exponential powers of feature interactions into simple linear combinations of the modulus and phase of the complex features, making it possible to adaptively learn the high-order feature interactions in an efficient way. Furthermore, EulerNet incorporates the implicit and explicit feature interactions into a unified architecture, which achieves the mutual enhancement and largely boosts the model capabilities. Such a network can be fully learned from data, with no need of pre-designed form or order for feature interactions. Extensive experiments conducted on three public datasets have demonstrated the effectiveness and efficiency of our approach. Our code is available at: https://github.com/RUCAIBox/EulerNet.