經典！工業界深度推薦系統與CTR預估必讀的論文匯總

• 2019 年 10 月 7 日
• 筆記

導讀：本文是「深度推薦系統」專欄的第十一篇文章，這個系列將介紹在深度學習的強力驅動下，給推薦系統工業界所帶來的最前沿的變化。本文主要根據Google推出的引領推薦系統與CTR預估工業界潮流至今的一大神作W&D[1]，所總結出來的深度推薦系統與CTR預估工業界必讀的論文匯總。 歡迎轉載，轉載請註明出處以及鏈接，更多關於深度推薦系統優質內容請關注如下頻道。 知乎專欄：深度推薦系統 微博：深度傳送門 公眾號：深度傳送門

起初是因為在唐傑老師的微博上看到其學生整理的一個關於Bert論文高引用相關的圖片（https://weibo.com/2126427211/I4cXHxIy4）。

一個偉大的學生做的一個BERT的論文以及它引用的文章之間的關係，相當於是一個針對論文Citation的Finding->Reasoning->Exploring的過程。感覺做得很酷，忍不住share出來了。。。他偉大的決定要寫個算法自動搞定！

覺得這個整理思路不錯，於是也照葫蘆畫瓢整理了一下推薦系統和CTR預估上工業界同樣鼎鼎大名Google出品的W&D[1]論文相關高引用的論文匯總。其實主要是對近年來推薦系統和CTR預估工業界影響力較大的論文做一個簡單的思路梳理，首先上圖如下，圓圈內數字為論文被引用數量。

Collaborative Filtering

• [WWW 17] Neural Collaborative Filtering
• [SIGIR 18] Collaborative Memory Network for Recommendation Systems

Deep部分演進

• [SIGIR 17] Neural Factorization Machines for Sparse Predictive Analytics
• [IJCAI 17] Attentional Factorization Machines: Learning the Weight of Feature Interactions via Attention Networks
• [ECIR 16] Factorization-supported Neural Network
• [TOIS 18] Product-Based Neural Networks for User Response Prediction over Multi-Field Categorical Data
• [RecSys 19] FiBiNET: Combining Feature Importance and Bilinear feature Interaction for Click-Through Rate Prediction
• [KDD 18] Deep Interest Network for Click-through Rate Prediction
• [AAAI 19] Deep Interest Evolution Network for Click-Through Rate Prediction
• [IJCAI 19] Deep Session Interest Network for Click-Through Rate Prediction
• [CIKM 19] AutoInt: Automatic Feature Interaction Learning via Self-Attentive Neural Networks

Wide部分演進

• [IJCAI 17] DeepFM: A Factorization-Machine based Neural Network for CTR Prediction
• [KDD 17] Deep & Cross Network for Ad Click Predictions
• [KDD 18] xDeepFM: Combining Explicit and Implicit Feature Interactions
• for Recommender Systems
• [WWW 19] Feature Generation by Convolutional Neural Network for Click-Through Rate Prediction

強化學習

• [WWW 17] DRN: A Deep Reinforcement Learning Framework for News Recommendation
• [WSDM 19] Top-K Off-Policy Correction for a REINFORCE Recommender System
• [IJCAI 19] Reinforcement Learning for Slate-based Recommender Systems: A Tractable Decomposition and Practical Methodology

知識圖譜

• [WWW 17] DKN: Deep Knowledge-Aware Network for News Recommendation
• [CIKM 18] RippleNet: Propagating User Preferences on the Knowledge Graph for Recommender Systems

Embedding技術

• [ICCCA 18] Item2Vec-Neural Item Embedding for Collaborative Filtering
• [RecSys 16] Meta-Prod2Vec: Product Embeddings Using Side-Information for Recommendation
• [KDD 18] Real-time Personalization using Embeddings for Search Ranking at Airbnb
• [KDD 18] Billion-scale Commodity Embedding for E-commerce Recommendation in Alibaba
• [WWW 19] NetSMF: Large-Scale Network Embedding as Sparse Matrix Factorization
• [IJCAI 19] ProNE: Fast and Scalable Network Representation Learning

參考文獻

[1] Wide & Deep Learning for Recommender Systems, DLRS 2016