Collaborative Deep Learning for Recommender Systems

By Hao Wang, Naiyan Wang, Dit-Yan Yeung. Knowledge Discovery and Data Mining, 2014

1,715 citations recorded by the source corpus.

Collaborative filtering (CF) is a successful approach commonly used by many recommender systems. Conventional CF-based methods use the ratings given to items by users as the sole source of information for learning to make recommendation. However, the ratings are often very sparse in many applications, causing CF-based methods to degrade significantly in their recommendation performance. To address this sparsity problem, auxiliary information such as item content information may be utilized. Collaborative topic regression (CTR) is an appealing recent method taking this approach which tightly couples the two components that learn from two different sources of information. Nevertheless, the latent representation learned by CTR may not be very effective when the auxiliary information is very sparse. To address this problem, we generalize recently advances in deep learning from i.i.d. input to non-i.i.d. (CF-based) input and propose in this paper a hierarchical Bayesian model called collaborative deep learning (CDL), which jointly performs deep representation learning for the content information and collaborative filtering for the ratings (feedback) matrix. Extensive experiments on three real-world datasets from different domains show that CDL can significantly advance the state of the art.

Subfields: Artificial Intelligence, Information Systems, Representation & Self-Supervised Learning

View Paper: https://arxiv.org/abs/1409.2944

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Xiangnan He
2. Yongfeng Zhang
3. Xu Chen
4. Xing Xie
5. Tat-seng Chua
6. Qiang Yang
7. Xiang Wang
8. Dacheng Tao

In Information Systems

1. Xiangnan He
2. Yongfeng Zhang
3. Xu Chen
4. Xing Xie
5. Xiang Wang
6. Tat-seng Chua
7. Naomie Binti Salim
8. Aminu Da’u

In Representation & Self-Supervised Learning

1. Xiangnan He
2. Qiang Yang
3. Yu Zhang
4. Yongfeng Zhang
5. Philip S. Yu
6. Tat-seng Chua
7. Xu Chen
8. Dit-Yan Yeung

Citations in the corpus

• On Sampling Strategies for Neural Network-based Collaborative Filtering (2017, methodology): “As pointed out in [32], the precision is not a suitable performance measure since non interactions may be due to (1) the user is not interested in the item, or (2) the user does not pay attention to its existence.”
• Variational Autoencoders for Collaborative Filtering (2018, background): “1), various approaches [2, 28, 47, 48] have applied neural networks to incorporate side information into collaborative filtering models to better handle the cold-start problem.”
• Recurrent Latent Variable Networks for Session-Based Recommendation (2017, methodology): “More recently, the method in [32] presented a hierarchical Bayesian model called collaborative deep learning (CDL) for RS.”
• Deep Neural Networks for YouTube Recommendations (2016, neutral): “Collaborative filtering is formulated as a deep neural network in [22] and autoencoders in [18].”
• Improved Recurrent Neural Networks for Session-based Recommendations (2016, methodology): “They have also been used for collaborative filtering [29, 21].”
• Session-based Recommendations with Recurrent Neural Networks (2015, neutral): “More recently Wang et al. (2015) introduced a more generic approach whereby a deep network is used to extract generic content-features from any types of items, these features are then incorporated in a standard collaborative filtering model to enhance the recommendation performance.”
• DKN: Deep Knowledge-Aware Network for News Recommendation (2018, methodology): “In addition to the aforementioned DSSM [16], DeepWide [6], DeepFM [13], YouTubeNet [8] and DMF [50], other popular deep-learning-based recommender systems include Collaborative Deep Learning [44], SHINE [45], Multi-view Deep Learning [11], and Neural Collaborative Filtering [14].”
• Joint Topic-Semantic-aware Social Recommendation for Online Voting (2017, methodology): “In addition, various models are incorporated into collaborative filtering, such as Support Vector Machine [30], Restricted Boltzmann Machine [24], and Stacked Denoising Auto Encoder [27].”
• Wide & Deep Learning for Recommender Systems (2016, background): “In the recommender systems literature, collaborative deep learning has been explored by coupling deep learning for content information and collaborative filtering (CF) for the ratings matrix [7].”
• Deep Learning for IoT Big Data and Streaming Analytics: A Survey (2017, background): “Moreover, DL can be used as a personalized recommendation module [142] to recommend more relevant content to the educator.”
• Knowledge-aware Autoencoders for Explainable Recommender Systems (2018, background): “In other works such as [22] a stacked architecture made of Autoencoders is proposed to perform a generalization over a higher set of latent features that every stacked autoencoder is able to learn.”
• Natural-Parameter Networks: A Class of Probabilistic Neural Networks (2016, background): “As possible extensions of NPN, it would be interesting to connect NPN to arbitrary PGM to form fully Bayesian deep learning models [24, 25], allowing even richer descriptions of relationships among latent variables.”
• DeepFM: A Factorization-Machine based Neural Network for CTR Prediction (2017, neutral): “[ Salakhutdinov et al. , 2007; Sedhain et al. , 2015; Wang et al. , 2015 ]”
• Joint Deep Modeling of Users and Items Using Reviews for Recommendation (2017, methodology): “A similar approach is followed in [35] for movie recommendation by using a generalized Stacked Auto Encoder (SAE) model.”
• A Hybrid Collaborative Filtering Model with Deep Structure for Recommender Systems (2017, methodology): “Similar to (Wang, Wang, and Yeung 2015; Wang and Blei 2011), we use recall as another evaluation metric since the rating information is in the form of implicit feedback (Hu, Koren, and Volinsky 2008; Rendle et al. 2009).”
• Bridging Language and Items for Retrieval and Recommendation: Benchmarking LLMs as Semantic Encoders (2024, neutral): “Early works in content-based recommendation (Mooney and Roy, 2000; Gopalan et al., 2014) extracted keywords and learned their collaborative relations between users and items (Ren-dle, 2010; Wang and Blei, 2011; Chen et al., 2012; Wang et al., 2015).”
• A Brief Review of Domain Adaptation (2020, background): “Some real-world application examples of multiview learning are video analysis [5], [6], speaker recognition [7], [8], natural language processing [9], [10], [11], and recommender system [12], [13].”