Masked Multi-Domain Network: Multi-Type and Multi-Scenario Conversion Rate Prediction with a Single Model

no code implementations • 26 Mar 2024 • Wentao Ouyang, Xiuwu Zhang, Chaofeng Guo, Shukui Ren, Yupei Sui, Kun Zhang, Jinmei Luo, Yunfeng Chen, Dongbo Xu, Xiangzheng Liu, Yanlong Du

A desired model for this problem should satisfy the following requirements: 1) Accuracy: the model should achieve fine-grained accuracy with respect to any conversion type in any display scenario.

Contrastive Learning for Conversion Rate Prediction

1 code implementation • 12 Jul 2023 • Wentao Ouyang, Rui Dong, Xiuwu Zhang, Chaofeng Guo, Jinmei Luo, Xiangzheng Liu, Yanlong Du

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.

Contrastive Learning

The Short Text Matching Model Enhanced with Knowledge via Contrastive Learning

no code implementations • 8 Apr 2023 • Ruiqiang Liu, Qiqiang Zhong, Mengmeng Cui, Hanjie Mai, Qiang Zhang, Shaohua Xu, Xiangzheng Liu, Yanlong Du

The model uses a generative model to generate corresponding complement sentences and uses the contrastive learning method to guide the model to obtain more semantically meaningful encoding of the original sentence.

Contrastive Learning Sentence +1

Learning Graph Meta Embeddings for Cold-Start Ads in Click-Through Rate Prediction

1 code implementation • 19 May 2021 • Wentao Ouyang, Xiuwu Zhang, Shukui Ren, Li Li, Kun Zhang, Jinmei Luo, Zhaojie Liu, Yanlong Du

For existing old ads, GMEs first build a graph to connect them with new ads, and then adaptively distill useful information.

Click-Through Rate Prediction Meta-Learning

MiNet: Mixed Interest Network for Cross-Domain Click-Through Rate Prediction

1 code implementation • 7 Aug 2020 • Wentao Ouyang, Xiuwu Zhang, Lei Zhao, Jinmei Luo, Yu Zhang, Heng Zou, Zhaojie Liu, Yanlong Du

Our study is based on UC Toutiao (a news feed service integrated with the UC Browser App, serving hundreds of millions of users daily), where the source domain is the news and the target domain is the ad.

Click-Through Rate Prediction

Click-Through Rate Prediction with the User Memory Network

1 code implementation • 9 Jul 2019 • Wentao Ouyang, Xiuwu Zhang, Shukui Ren, Li Li, Zhaojie Liu, Yanlong Du

Both offline and online experiments demonstrate the effectiveness of MA-DNN for practical CTR prediction services.

Click-Through Rate Prediction

Representation Learning-Assisted Click-Through Rate Prediction

1 code implementation • 11 Jun 2019 • Wentao Ouyang, Xiuwu Zhang, Shukui Ren, Chao Qi, Zhaojie Liu, Yanlong Du

These subnets model the user-ad, ad-ad and feature-CTR relationship respectively.

Click-Through Rate Prediction Representation Learning

Deep Spatio-Temporal Neural Networks for Click-Through Rate Prediction

1 code implementation • 10 Jun 2019 • Wentao Ouyang, Xiuwu Zhang, Li Li, Heng Zou, Xin Xing, Zhaojie Liu, Yanlong Du

The intuitions are that ads shown together may influence each other, clicked ads reflect a user's preferences, and unclicked ads may indicate what a user dislikes to certain extent.

Click-Through Rate Prediction