Visual Encoding and Debiasing for CTR Prediction

TL;DR

This paper introduces a contrastive learning-based visual encoding framework for CTR prediction that produces fine-grained, unbiased visual features, improving accuracy in large-scale online advertising systems.

Contribution

The paper proposes a novel contrastive learning approach with click supervision and a debiasing network to enhance visual feature quality and reduce bias in CTR prediction.

Findings

Improved CTR prediction accuracy in offline experiments.

Reduced bias in visual features through debiasing techniques.

Enhanced online performance in Alibaba's visual search system.

Abstract

Extracting expressive visual features is crucial for accurate Click-Through-Rate (CTR) prediction in visual search advertising systems. Current commercial systems use off-the-shelf visual encoders to facilitate fast online service. However, the extracted visual features are coarse-grained and/or biased. In this paper, we present a visual encoding framework for CTR prediction to overcome these problems. The framework is based on contrastive learning which pulls positive pairs closer and pushes negative pairs apart in the visual feature space. To obtain fine-grained visual features,we present contrastive learning supervised by click through data to fine-tune the visual encoder. To reduce sample selection bias, firstly we train the visual encoder offline by leveraging both unbiased self-supervision and click supervision signals. Secondly, we incorporate a debiasing network in the online CTR predictor to adjust the visual features by contrasting high impression items with selected items with lower impressions.We deploy the framework in the visual sponsor search system at Alibaba. Offline experiments on billion-scale datasets and online experiments demonstrate that the proposed framework can make accurate and unbiased predictions.