Collaborative Denoising Auto-Encoders for Top-N Recommender Systems

By Yao Wu, Christopher DuBois, Alice X. Zheng, Martin Ester. 2016

1,053 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Information Systems, Representation & Self-Supervised Learning, Recommender Systems

How researchers see this paper

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

In Computer Vision and Pattern Recognition

1. Xiangnan He
2. Tat-seng Chua
3. Xiang Wang
4. Yongfeng Zhang
5. Xu Chen
6. Fuli Feng
7. Liqiang Nie
8. Qiang Yang

In Information Systems

1. Xiangnan He
2. Xiang Wang
3. Tat-seng Chua
4. Yongfeng Zhang
5. Xu Chen
6. Meng Wang
7. Fuli Feng
8. Xing Xie

In Representation & Self-Supervised Learning

1. Xiangnan He
2. Tat-seng Chua
3. Xiang Wang
4. Yongfeng Zhang
5. Qiang Yang
6. Xu Chen
7. Fuli Feng
8. Liqiang Nie

In Recommender Systems

1. Xiangnan He
2. Tat-seng Chua
3. Yongfeng Zhang
4. Xiang Wang
5. Xu Chen
6. Harald Steck
7. Fuli Feng
8. Xing Xie

Citations in the corpus

• Variational Autoencoders for Collaborative Filtering (2018, methodology): “lix dataset, the parallelized grid search took about two weeks). We also found that the faster approximation of Slim [26] did not yield competitive performance. Collaborativedenoisingautoencoder(cdae)[51]:augments the standard denoising autoencoder by adding a per-user latent factor to the input. We”
• On the challenges of learning with inference networks on sparse, high-dimensional data (2017, background): “With them, the same generative model , outperforms CDAE and WMF on both datasets, and marginally outperforms SLIM on ML-20M while achieving nearly state-of-the-art re-sults on Netflix.”
• Neural Collaborative Filtering (2017, methodology): “This may partially explain why using deep layers for CDAE does not improve the performance ( cf. Section 6 of [44]).”
• Explainable Recommendation: A Survey and New Perspectives (2018, methodology): “Though the term explainable recommendation was formally introduced in the recent years (Zhang et al. [2014a]), the basic concept, however, dates back to some of the most early works in personalized recommendation research.”
• Bridging Collaborative Filtering and Semi-Supervised Learning: A Neural Approach for POI Recommendation (2017, positive): “Recently, autoencoders have also been successfully adopted for learning user representations based on rated items [23, 36, 44].”
• Joint Representation Learning for Top-N Recommendation with Heterogeneous Information Sources (2017, methodology): “For example, [25, 38] adopted denoising auto-encoders for recommendation, [46] developed a neural autoregressive approach for collaborative €ltering, and [19, 20] generalized matrix factorization and factorization machines for neural collaborative €ltering.”
• xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systems (2018, methodology): “[35, 41] model CF base on the autoencoder paradigm, and they have empirically demonstrated that autoencoder-based CF outperforms several classical MF models.”
• Embarrassingly Shallow Autoencoders for Sparse Data (2019, background): “While the area of collaborative filtering has long been dominated by matrix factorization approaches, recent years have witnessed a surge in deep learning approaches [5, 7–9, 13, 21, 25, 26], spurred by their great successes in other fields.”
• Bias and Debias in Recommender System: A Survey and Future Directions (2020, methodology): “For example, Logistical matrix factorization [78], BPR [138], or most of neural-based recommendation models (e.g. CDAE [184], NCF [64], LightGCN [63]) apply the uniform negative sampler; Yu et al. [194] considers to over-sample the popular negative items, as they are more likely to be exposed.”
• Hypergraph Contrastive Collaborative Filtering (2022, positive): “The hyperbolic embedding space is adopted to encode high-order information from neighboring users/items in [32].”
• Neural Graph Collaborative Filtering (2019, background): “f an observed interaction close to each other, its linearity makes it insufficient to reveal the complex and nonlinear relationships between users and items [14, 15]. Towards this end, recent efforts [11, 14, 15, 35] focus on exploiting deep learning techniques to enhance the interaction function, s”
• NAIS: Neural Attentive Item Similarity Model for Recommendation (2018, methodology): “, images and texts [39], [40], [41], [42], [43], and 2) using DNNs for learning the useritem scoring function [19], [44], [45], [46].”
• Computing Recommendations via a Knowledge Graph-aware Autoencoder (2018, methodology): “In [47], the authors propose the usage of a denosing autoencoder to perform a top-N recommendation task.”
• Recommendation system based on deep learning methods: a systematic review and new directions (2019, methodology): “This approach was later extended by incorporating side information, item description and user profile to address the problem of cold start (Strub et al. 2016a; Wu et al. 2016b, c).”
• Neural Attentive Session-based Recommendation (2017, methodology): “Recently, denoising auto-encoders have been used to perform CF in a similar manner [34, 43].”
• Deep Item-based Collaborative Filtering for Top-N Recommendation (2018, methodology): “Moreover, several prior efforts [8, 9, 43] provide empirical evidences on accuracy superiority of ICF over UCF.”
• Denoising Implicit Feedback for Recommendation (2020, background): “te the robustness of recommender systems under various attack methods, such as shilling attacks [17] and fuzzing attacks [29]. To build more robust recommender systems, some auto-encoder based models [20, 30, 33] introduce the denoising techniques. These approaches [33] firstcorrupttheinput,i.e.,the”