版权声明：转载请注明出处，谢谢！！！ https://blog.csdn.net/tszupup/article/details/84968503
最近读了2018年发表在ACM Computing Snrveys上的一篇关于基于深度学习的推荐系统的比较全面的综述《Deep Learning based Recommender System: A Survey and New Perspectives》，这篇综述简要介绍了多种常用的深度学习技术及其在推荐领域的应用，每种深度学习模型对应多篇参考文献，一共200多篇文献，从每种深度学习模型或框架对应的参考文献中挑出了几篇，如下：
1. 多层感知机/MLP
- A multi-view deep learning approach for cross domain user modeling in recommendation systems
- Learning deep structured semantic models for web search using clickthrough data
- Neural Network Matrix Factorization
- Wide & deep learning for recommender systems
2. 自编码器/AE
- Autorec: Autoencoders meet collaborative filltering
- AutoSVD++: An Efficient Hybrid Collaborative Filtering Model via Contractive Auto-encoders
- Collaborative deep learning for recommender systems
- Collaborative deep ranking: a hybrid pair-wise recommendation algorithm with implicit feedback
- Collaborative denoising auto-encoders for top-n recommender systems
- Deep collaborative €filtering via marginalized denoising auto-encoder
- DLTSR: A Deep Learning Framework for Recommendation of Long-tail Web Services
3. 卷积神经网络/CNN
- 3D Convolutional Networks for Session-based Recommendation with Content Features
- Collaborative Deep Metric Learning for Video Understanding
- ConTagNet: exploiting user context for image tag recommendation
- Convolutional matrix factorization for document context-aware recommendation
- Deep content-based music recommendation
- Graph convolutional matrix completion
- Graph Convolutional Neural Networks for Web-Scale Recommender Systems
- Joint Deep Modeling of Users and Items Using Reviews for Recommendation
- Outer Product-based Neural Collaborative Filtering
- Personalized top-n sequential recommendation via convolutional sequence embedding
- What Your Images Reveal: Exploiting Visual Contents for Point-of-Interest Recommendation
4. 循环神经网络/RNN
- Contextual Sequence Modeling for Recommendation with Recurrent Neural Networks
- Joint Deep Modeling of Users and Items Using Reviews for Recommendation
- Multi-rate deep learning for temporal recommendation
- Neural Rating Regression with Abstractive Tips Generation for Recommendation
- Parallel recurrent neural network architectures for feature-rich session-based recommendation
- Personal recommendation using deep recurrent neural networks in NetEase
- Q&R: A Two-Stage Approach toward Interactive
- Recurrent coevolutionary latent feature processes for continuous-time recommendation
- Recurrent recommender networks
- Session-based recommendations with recurrent neural networks
5. 受限玻尔兹曼机/RBM
- A non-iid framework for collaborative filltering with restricted boltzmann machines
- atrank: an attention-based user behavior modeling framework for recommend
- Restricted Boltzmann machines for collaborative filtering
6. 神经自回归分布估计/NADE
- A Neural Autoregressive Approach to Collaborative Filtering
- Collaborative Filtering with User-Item Co-Autoregressive Models
7. 对抗网络/AN
- Adversarial Personalized Ranking for Recommendation
8. 注意力模型/AM
- Hashtag Recommendation for Multimodal Microblog Using Co-Attention Network
- Hashtag recommendation using attention-based convolutional neural network
- Hashtag recommendation with topical attention-based LSTM
- Next Item Recommendation with Self-Attention
- Sequential Recommender System based on Hierarchical Attention Networks
9. 深度强化学习/DRL
- A Contextual-Bandit Approach to Personalized News Article Recommendation
- DRN: A Deep Reinforcement Learning Framework for News Recommendation
- IRGAN: A Minimax Game for Unifying Generative and Discriminative Information Retrieval Models
- Recommendations with Negative Feedback via Pairwise Deep Reinforcement Learning
10. CNN & Autoencoder
- Collaborative Knowledge Base Embedding forRecommender Systems
11. CNN & Autoencoder
- Hashtag Recommendation for Multimodal MicroblogUsing Co-Attention Network
- Neural Citation Network for Context-Aware CitationRecommendation
- Personalized Key Frame Recommendation
- Quote Recommendation in Dialogueusing Deep Neural Network
12. RNN & Autoencoder
- Collaborative Recurrent Autoencoder–Recommend while Learning to Fill in the Blanks
13. RNN with DRL
- Supervised Reinforcement Learning with Recurrent Neural Network for Dynamic Treatment Recommendation
欢迎交流！ QQ：3408649893