Entire Space Multi-Task Model: An Effective Approach for Estimating Post-Click Conversion Rate

Abstract

Estimating post-click conversion rate (CVR) accurately is crucial for ranking systems in industrial applications such as recommendation and advertising. Conventional CVR modeling applies popular deep learning methods and achieves state-of-the-art performance. However it encounters several task-specific problems in practice, making CVR modeling challenging. For example, conventional CVR models are trained with samples of clicked impressions while utilized to make inference on the entire space with samples of all impressions. This causes a sample selection bias problem. Besides, there exists an extreme data sparsity problem, making the model fitting rather difficult. In this paper, we model CVR in a brand-new perspective by making good use of sequential pattern of user actions, i.e., impression -> click -> conversion. The proposed Entire Space Multi-task Model (ESMM) can eliminate the two problems simultaneously by i) modeling CVR directly over the entire space, ii) employing a feature representation transfer learning strategy. Experiments on dataset gathered from Taobao's recommender system demonstrate that ESMM significantly outperforms competitive methods. We also release a sampling version of this dataset to enable future research. To the best of our knowledge, this is the first public dataset which contains samples with sequential dependence of click and conversion labels for CVR modeling.

arXiv commentaccept by SIGIR-2018

References in corpus (2)

• An Overview of Multi-Task Learning in Deep Neural Networks
• Deep Interest Network for Click-Through Rate Prediction

Cited by in corpus (26)

• Lifelong Sequential Modeling with Personalized Memorization for User Response Prediction
• KuaiRand: An Unbiased Sequential Recommendation Dataset with Randomly Exposed Videos
• ESCM${}^2$: Entire Space Counterfactual Multi-Task Model for Post-Click Conversion Rate Estimation
• SAR-Net: A Scenario-Aware Ranking Network for Personalized Fair Recommendation in Hundreds of Travel Scenarios
• Enhanced Doubly Robust Learning for Debiasing Post-click Conversion Rate Estimation
• AdaTask: A Task-aware Adaptive Learning Rate Approach to Multi-task Learning
• MetaBalance: Improving Multi-Task Recommendations via Adapting Gradient Magnitudes of Auxiliary Tasks
• Multi-Task Recommendations with Reinforcement Learning
• Automatic Expert Selection for Multi-Scenario and Multi-Task Search
• Balancing Unobserved Confounding with a Few Unbiased Ratings in Debiased Recommendations
• DESCN: Deep Entire Space Cross Networks for Individual Treatment Effect Estimation
• Scenario-Adaptive and Self-Supervised Model for Multi-Scenario Personalized Recommendation
• A Multi-task Learning Framework for Product Ranking with BERT
• Multi-Scenario Ranking with Adaptive Feature Learning
• Mixture of Virtual-Kernel Experts for Multi-Objective User Profile Modeling
• EdgeRec: Recommender System on Edge in Mobile Taobao
• Entire Space Multi-Task Modeling via Post-Click Behavior Decomposition for Conversion Rate Prediction
• CPMR: Context-Aware Incremental Sequential Recommendation with Pseudo-Multi-Task Learning
• GreenFlow: A Computation Allocation Framework for Building Environmentally Sound Recommendation System
• DADNN: Multi-Scene CTR Prediction via Domain-Aware Deep Neural Network
• Deep Bayesian Multi-Target Learning for Recommender Systems
• Rethinking Large-scale Pre-ranking System: Entire-chain Cross-domain Models
• NEON: Living Needs Prediction System in Meituan
• POSGen: Personalized Opening Sentence Generation for Online Insurance Sales
• Practical Compositional Fairness: Understanding Fairness in Multi-Component Recommender Systems
• Modeling Orders of User Behaviors via Differentiable Sorting: A Multi-task Framework to Predicting User Post-click Conversion