Deep Recommenders

Deep Recommenders is an open-source recommendation system algorithm library built by tf.estimator and tf.keras that the advanced APIs of TensorFlow.

🤗️ This Library mainly used for self-learning and improvement, but also hope to help friends and classmates who are interested in the recommendation system to make progress together!

Models

Ranking

• FM [Estimator] _{Factorization Machines, Osaka, 2010}
• FFM _{Field-aware Factorization Machines for CTR Prediction, RecSys, 2016}
• LS-PLM _{Learning Piece-wise Linear Models from Large Scale Data for Ad Click Prediction, Alibaba, 2017}
• WDL [Estimator] _{Wide & Deep Learning for Recommender Systems, Google, DLRS, 2016}
• PNN _{Product-based Neural Networks for User Response Prediction, IEEE, 2016}
• FNN [Estimator] _{Deep Learning over Multi-field Categorical Data: A Case Study on User Response Prediction, RayCloud, ECIR, 2016}
• NFM _{Neural Factorization Machines for Sparse Predictive Analytics, SIGIR, 2017}
• AFM _{Attentional Factorization Machines: Learning the Weight of Feature Interactions via Attention Networks, IJCAI, 2017}
• DeepFM [Estimator] [Keras] _{DeepFM: A Factorization-Machine based Neural Network for CTR Prediction, Huawei, IJCAI, 2017}
• DCN _{Deep & Cross Network for Ad Click Predictions, Google, KDD, 2017}
• xDeepFM _{xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systems, Microsoft, KDD, 2018}
• DIN _{Deep Interest Network for Click-Through Rate Prediction, Alibaba, KDD, 2018}
• DIEN _{Deep Interest Evolution Network for Click-Through Rate Prediction, Alibaba, AAAI, 2019}
• DLRM _{Deep Learning Recommendation Model for Personalization and Recommendation Systems, Facebook, 2019}

Retrieval

• DSSM _{Learning Deep Structured Semantic Models for Web Search using Clickthrough Data, Microsoft, CIKM, 2013}
• YoutubeNet _{Deep Neural Networks for YouTube Recommendations, Google, RecSys, 2016}
• SBCNM _{Sampling-Bias-Corrected Neural Modeling for Large Corpus Item Recommendations, Google, RecSys, 2019}
• EBR _{Embedding-based Retrieval in Facebook Search, Facebook, KDD, 2020}
• Item2Vec _{Item2Vec: Neural Item Embedding for Collaborative Filtering, Microsoft, MLSP, 2016}
• Airbnb _{Real-time Personalization using Embeddings for Search Ranking at Airbnb, Airbnb, KDD, 2018}
• DeepWalk _{DeepWalk: Online Learning of Social Representations, StonyBrook, KDD, 2014}
• EGES _{Billion-scale Commodity Embedding for E-commerce Recommendation in Alibaba, Alibaba, KDD, 2018}
• GCN [Keras] _{Semi-Supervised Classification with Graph Convolutional Networks, ICLR, 2017}
• GraphSAGE _{Inductive Representation Learning on Large Graphs, NIPS, 2017}
• PinSage _{Graph Convolutional Neural Networks for Web-Scale Recommender Systems, Pinterest, KDD, 2018}
• IntentGC _{IntentGC: a Scalable Graph Convolution Framework Fusing Heterogeneous Information for Recommendation, Alibaba, KDD, 2019}
• GraphTR _{Graph Neural Network for Tag Ranking in Tag-enhanced Video Recommendation, Tencent, CIKM, 2020}

Multi-task learning

• MMoE [Estimator] _{Modeling Task Relationships in Multi-task Learning with Multi-gate Mixture-of-Experts, Google, KDD, 2018}
• ESMM _{Entire Space Multi-Task Model: An Effective Approach for Estimating Post-Click Conversion Rate, Alibaba, SIGIR, 2018}

NLP

• Word2Vec _{Distributed Representations of Words and Phrases and their Compositionality, Google, NIPS, 2013}

• Transformer [Keras] _{Attention Is All You Need, Google, NeurlPS, 2017}

• BERT _{BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding, Google, NAACL, 2019}

Supports

Modules	TensorFlow
deep_recommenders.estimator
deep_recommenders.keras

License

Apache License 2.0