Category-Specific CNN for Visual-aware CTR Prediction at JD.com

Abstract

As one of the largest B2C e-commerce platforms in China, JD.com also powers a leading advertising system, serving millions of advertisers with fingertip connection to hundreds of millions of customers. In our system, as well as most e-commerce scenarios, ads are displayed with images. This makes visual-aware Click Through Rate (CTR) prediction of crucial importance to both business effectiveness and user experience. Existing algorithms usually extract visual features using off-the-shelf Convolutional Neural Networks (CNNs) and late fuse the visual and non-visual features for the finally predicted CTR. Despite being extensively studied, this field still face two key challenges. First, although encouraging progress has been made in offline studies, applying CNNs in real systems remains non-trivial, due to the strict requirements for efficient end-to-end training and low-latency online serving. Second, the off-the-shelf CNNs and late fusion architectures are suboptimal. Specifically, off-the-shelf CNNs were designed for classification thus never take categories as input features. While in e-commerce, categories are precisely labeled and contain abundant visual priors that will help the visual modeling. Unaware of the ad category, these CNNs may extract some unnecessary category-unrelated features, wasting CNN's limited expression ability. To overcome the two challenges, we propose Category-specific CNN (CSCNN) specially for CTR prediction. CSCNN early incorporates the category knowledge with a light-weighted attention-module on each convolutional layer. This enables CSCNN to extract expressive category-specific visual patterns that benefit the CTR prediction. Offline experiments on benchmark and a 10 billion scale real production dataset from JD, together with an Online A/B test show that CSCNN outperforms all compared state-of-the-art algorithms. We also build a highly efficient infrastructure to accomplish end-to-end training with CNN

A Hybrid Bandit Model with Visual Priors for Creative Ranking in Display Advertising

“…CTR prediction of image ads is a core task of online display advertising systems. Due to the recent advances in computer vision, visual features are employed to further enhance the recommendation models [3,6,8,9,14,20,21,31,33]. [3,9] quantitatively study the relationship between handcrafted visual features and creative online performance.…”

“…[8,21,31] extend these methods by training the CNNs in an end-to-end manner. [20] integrate the category information on top of the CNN embedding to help visual modeling. The above works focus on improving the product ranking by considering visual information while neglecting the great potential of creative ranking.…”

Hybrid CNN Based Attention with Category Prior for User Image Behavior Modeling

“…An off-the-shelf CNN is used in the two-stage structure, which results in a suboptimal model. Researches [1,12] on ad image representation learning have verified that the end-to-end methods learn better representation than the two-stage methods. However, the end-to-end method is not feasible for modeling user behavior image.…”

“…[13] shows that it is useful to consider ad category when extracting the representations of ad images. [12] shows that using a pretrained CNNs to extract embeddings for ad images is sub-optimal and incorporating the ad category into CNNs can improve the representing.…”

Automated Creative Optimization for E-Commerce Advertising

“…The Creatives are ubiquitous for online advertisements. Some works have paid attention to CTR prediction on ad creatives [5,26,29,39,42] via extracting expressive visual features to increase the CTR. However, there are few works about the optimization for advertising creatives given limited feedbacks.…”