IRLGMLJul 1, 2018

Product-based Neural Networks for User Response Prediction over Multi-field Categorical Data

arXiv:1807.00311v1210 citations
Originality Incremental advance
AI Analysis

This work improves personalized information retrieval and filtering, such as in recommender systems and web search, by enhancing prediction accuracy over multi-field categorical data, though it is incremental as it builds on existing neural network approaches.

The paper tackles user response prediction for click prediction by addressing gradient issues in existing models, proposing Product-based Neural Networks (PNN) and Product-network In Network (PIN) to learn feature interactions, resulting in consistent outperformance of 8 baselines on AUC and log loss across 5 datasets and a 34.67% relative CTR improvement in online A/B tests.

User response prediction is a crucial component for personalized information retrieval and filtering scenarios, such as recommender system and web search. The data in user response prediction is mostly in a multi-field categorical format and transformed into sparse representations via one-hot encoding. Due to the sparsity problems in representation and optimization, most research focuses on feature engineering and shallow modeling. Recently, deep neural networks have attracted research attention on such a problem for their high capacity and end-to-end training scheme. In this paper, we study user response prediction in the scenario of click prediction. We first analyze a coupled gradient issue in latent vector-based models and propose kernel product to learn field-aware feature interactions. Then we discuss an insensitive gradient issue in DNN-based models and propose Product-based Neural Network (PNN) which adopts a feature extractor to explore feature interactions. Generalizing the kernel product to a net-in-net architecture, we further propose Product-network In Network (PIN) which can generalize previous models. Extensive experiments on 4 industrial datasets and 1 contest dataset demonstrate that our models consistently outperform 8 baselines on both AUC and log loss. Besides, PIN makes great CTR improvement (relatively 34.67%) in online A/B test.

Code Implementations10 repos

Data from Papers with Code (CC-BY-SA-4.0)

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes