Click Through Rate Prediction

Click-through rate (CTR) prediction aims to accurately estimate the probability of a user clicking on a recommended item, crucial for optimizing online advertising and recommender systems. Current research emphasizes efficient modeling of high-order feature interactions, often employing deep learning architectures like factorization machines, transformers, and neural networks with specialized modules for handling diverse data distributions and user behaviors (e.g., incorporating contextual information, lifelong user history, and multimodal data). These advancements improve prediction accuracy and efficiency, impacting the effectiveness of personalized recommendations and online advertising strategies. Furthermore, research explores techniques to address challenges like data sparsity, cold-start problems, and computational limitations in large-scale deployments.

Papers

December 21, 2021

Adversarial Gradient Driven Exploration for Deep Click-Through Rate Prediction
Kailun Wu, Zhangming Chan, Weijie Bian, Lejian Ren, Shiming Xiang, Shuguang Han, Hongbo Deng, Bo Zheng
Click Through Rate Prediction Active Exploration Exploration Strategy Large Scale Recommender System Generative Adversarial Exploration

December 7, 2021

Enhanced Exploration in Neural Feature Selection for Deep Click-Through Rate Prediction Models via Ensemble of Gating Layers
Lin Guan, Xia Xiao, Ming Chen, Youlong Cheng
Neural Architecture Search Feature Selection Click Through Rate Prediction Active Exploration Click Through Rate Gating Mechanism

December 6, 2021

A General Framework for Debiasing in CTR Prediction
Wenjie Chu, Shen Li, Chao Chen, Longfei Xu, Hengbin Cui, Kaikui Liu
General Framework Click Through Rate Prediction Self Debiasing Click Through Rate Debiasing Framework Click Model

November 9, 2021

Dynamic Parameterized Network for CTR Prediction
Jian Zhu, Congcong Liu, Pei Wang, Xiwei Zhao, Guangpeng Chen, Junsheng Jin, Changping Peng, Zhangang Lin, Jingping Shao
Click Through Rate Prediction Parameterized Network

Click Through Rate Prediction

Papers

Adversarial Gradient Driven Exploration for Deep Click-Through Rate Prediction

Enhanced Exploration in Neural Feature Selection for Deep Click-Through Rate Prediction Models via Ensemble of Gating Layers

A General Framework for Debiasing in CTR Prediction

Dynamic Parameterized Network for CTR Prediction