Contrastive Learning for Conversion Rate Prediction

TL;DR

This paper introduces CL4CVR, a contrastive learning framework that enhances conversion rate prediction by leveraging unlabeled data and addressing data sparsity through novel embedding masking, false negative elimination, and positive inclusion techniques.

Contribution

The paper proposes a novel contrastive learning framework for CVR prediction that effectively utilizes unlabeled data and tackles data sparsity with new embedding masking, false negative elimination, and positive inclusion methods.

Findings

CL4CVR outperforms baseline models on real-world datasets.

Embedding masking improves data augmentation for CVR prediction.

False negative elimination enhances contrastive learning effectiveness.

Abstract

Conversion rate (CVR) prediction plays an important role in advertising systems. Recently, supervised deep neural network-based models have shown promising performance in CVR prediction. However, they are data hungry and require an enormous amount of training data. In online advertising systems, although there are millions to billions of ads, users tend to click only a small set of them and to convert on an even smaller set. This data sparsity issue restricts the power of these deep models. In this paper, we propose the Contrastive Learning for CVR prediction (CL4CVR) framework. It associates the supervised CVR prediction task with a contrastive learning task, which can learn better data representations exploiting abundant unlabeled data and improve the CVR prediction performance. To tailor the contrastive learning task to the CVR prediction problem, we propose embedding masking (EM), rather than feature masking, to create two views of augmented samples. We also propose a false negative elimination (FNE) component to eliminate samples with the same feature as the anchor sample, to account for the natural property in user behavior data. We further propose a supervised positive inclusion (SPI) component to include additional positive samples for each anchor sample, in order to make full use of sparse but precious user conversion events. Experimental results on two real-world conversion datasets demonstrate the superior performance of CL4CVR. The source code is available at https://github.com/DongRuiHust/CL4CVR.