1. 经典论文
1.1 传统推荐算法
Collaborative Filtering (CF) - 协同过滤算法
- Using collaborative filtering to weave an information tapestry 【1992年】
https://dl.acm.org/doi/10.1145/138859.138867 - Amazon.com recommendations: item-to-item collaborative filtering 【2003年】
https://ieeexplore.ieee.org/abstract/document/1167344
Matrix factorization (MF) - 矩阵分解算法
- Matrix Factorization Techniques for Recommender Systems 【2009年】
https://ieeexplore.ieee.org/abstract/document/5197422
Logistic regression (LR) - 逻辑回归算法
Factorization machines (FM) - 因子分解机
- Factorization Machines 【2010年】
https://ieeexplore.ieee.org/abstract/document/5694074 - Field-aware Factorization Machines for CTR Prediction 【FFM算法，2016年】
https://ieeexplore.ieee.org/abstract/document/5694074
GBDT+LR （Facebook）
- Practical Lessons from Predicting Clicks on Ads at Facebook 【2014年】
https://dl.acm.org/doi/abs/10.1145/2648584.2648589
LS-PLM （阿里巴巴）
- Learning Piece-wise Linear Models from Large Scale Data for Ad Click Prediction 【2017年】
https://arxiv.org/abs/1704.05194
1.2 深度学习推荐算法
AutoRec - 单隐层神经网络推荐模型
- AutoRec: Autoencoders Meet Collaborative Filtering
http://users.cecs.anu.edu.au/~akmenon/papers/autorec/autorec-paper.pdf
Deep Crossing - 经典的深度学习架构
- Deep Crossing: Web-Scale Modeling without Manually Crafted Combinatorial Features
https://dl.acm.org/doi/abs/10.1145/2939672.2939704
Neural collaborative filtering （NCF）- CF与深度学习的结合
- Neural Collaborative Filtering
https://dl.acm.org/doi/abs/10.1145/3038912.3052569
PNN - 加强特征交叉能力
- Product-Based Neural Networks for User Response Prediction
https://ieeexplore.ieee.org/abstract/document/7837964
Wide & Deep - 记忆能力与泛化能力的综合
- Wide & Deep Learning for Recommender Systems 【Wide & Deep 模型】
https://dl.acm.org/doi/abs/10.1145/2988450.2988454 - Deep & Cross Network for Ad Click Predictions 【Deep & Cross模型】
https://arxiv.org/abs/1708.05123
FM与深度学习的结合
- Deep Learning over Multi-field Categorical Data【FNN模型】
https://link.springer.com/chapter/10.1007/978-3-319-30671-1_4 - DeepFM: A Factorization-Machine based Neural Network for CTR Prediction 【DeepFM模型】
https://www.ijcai.org/Proceedings/2017/0239.pdf - Neural Factorization Machines for Sparse Predictive Analytics【NFM模型】
https://arxiv.org/abs/1708.05027
注意力机制在推荐模型中的应用
- Attentional Factorization Machines: Learning the Weight of Feature Interactions via Attention Networks 【AFM模型】
https://arxiv.org/abs/1708.04617 - Deep Interest Network for Click-Through Rate Prediction 【DIN模型】
https://arxiv.org/abs/1706.06978
序列模型与推荐系统的结合
- Deep Interest Evolution Network for Click-Through Rate Prediction 【DIEN模型】
https://arxiv.org/abs/1809.03672
强化学习与推荐系统的结合
- DRN: A Deep Reinforcement Learning Framework for News 【DRN模型】
Recommendation
https://www.personal.psu.edu/~gjz5038/paper/www2018_reinforceRec/www2018_reinforceRec.pdf
1.3 双塔召回模型
- Deep Neural Networks for YouTube Recommendations
https://static.googleusercontent.com/media/research.google.com/zh-CN//pubs/archive/45530.pdf
2. 代码集合
- DeepCTR tensorflow：https://github.com/shenweichen/DeepCTR
