AutoRec: Autoencoders Meet Collaborative Filtering

By Suvash Sedhain, Aditya Krishna Menon, Scott Sanner, Lexing Xie. 2015

1,281 citations recorded by the source corpus.

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

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. Dacheng Tao
3. Luc Van Gool
4. Tat-seng Chua
5. Trevor Darrell
6. Xiaogang Wang
7. Jitendra Malik
8. Alan L. Yuille

In Information Systems

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

In Representation & Self-Supervised Learning

1. Xiangnan He
2. Tat-seng Chua
3. Xiang Wang
4. Michael M. Bronstein
5. Meng Wang
6. Xing Xie
7. Jiaxi Tang
8. Charu C. Aggarwal

In Recommender Systems

1. Xiangnan He
2. Tat-seng Chua
3. Xiang Wang
4. Xing Xie
5. Aminu Da’u
6. Naomie Binti Salim
7. Meng Wang
8. Liqiang Nie

In Signal Processing

1. Charu C. Aggarwal
2. Yu Zheng
3. Xiangnan He
4. Xing Xie
5. Julian McAuley
6. Jiawei Han
7. Naomie Binti Salim
8. Aminu Da’u

Citations in the corpus

• Variational Autoencoders for Collaborative Filtering (2018, background): “Recently, a growing body of work involves applying neural networks to the collaborative filtering setting with promising results [14, 41, 51, 54].”
• Neural Collaborative Filtering (2017, background): “Recently, autoencoders have become a popular choice for building recommendation systems [32, 22, 35].”
• Deep Neural Networks for YouTube Recommendations (2016, neutral): “Collaborative filtering is formulated as a deep neural network in [22] and autoencoders in [18].”
• A Neural Autoregressive Approach to Collaborative Filtering (2016, methodology): “We use a single hidden layer architecture and the number of hidden units is set to 500 , same as AutoRec (Sedhain et al., 2015) and LLORMA (Lee et al., 2013).”
• Knowledge-aware Autoencoders for Explainable Recommender Systems (2018, methodology): “Among them, Autoencoder Neural Networks have proven their effectiveness in CF settings as shown in [14], in which the authors use an Autoencoder fed with user ratings in order to predict missing value for users’ unseen items.”
• AutoEmb: Automated Embedding Dimensionality Search in Streaming Recommendations (2020, methodology): “Suvash et al.[27] introduced AutoRec along with item-based and user-based one to learn the lower-dimension feature representations of users and items.”
• Geometric Matrix Completion with Recurrent Multi-Graph Neural Networks (2017, background): “Recent works applying deep learning to recommender systems used networks with fully connected or auto-encoder architectures [44, 35, 14].”
• Ask the GRU: Multi-task Learning for Deep Text Recommendations (2016, background): “Recently, [37, 38] proposed de-noising auto-encoder based models for collaborative filter-ing which are trained to denoise corrupted versions of entire sparse vectors of user-item likes or item-user likes (i.e. rows or columns of the R matrix).”
• DeepFM: A Factorization-Machine based Neural Network for CTR Prediction (2017, background): “[ Salakhutdinov et al. , 2007; Sedhain et al. , 2015; Wang et al. , 2015 ]”
• Personalized Top-N Sequential Recommendation via Convolutional Sequence Embedding (2018, background): “Auto-encoder framework [25, 29] and its variant denoising autoencoder [32] also produce a good recommendation performance.”
• BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer (2019, methodology): “place of conventional matrix factorization. For example, Neural Collaborative Filtering (NCF) [12] estimates user preferences via Multi-Layer Perceptions (MLP) instead of inner product, while AutoRec [44] and CDAE [57] predict users’ ratings using Auto-encoder framework. 2.2 Sequential Recommendatio”
• Collaborative Filtering with Stacked Denoising AutoEncoders and Sparse Inputs (2015, methodology): “Yet, we also reported NL-PMF [8] as it is a non-linear algorithm and AutoRec [15] as it is based on neural networks.”
• Auto-Encoding User Ratings via Knowledge Graphs in Recommendation Scenarios (2017, methodology): “A pure CF approach has been settled with autoencoders in [23].”
• Deep Learning Recommendation Model for Personalization and Recommendation Systems (2019, methodology): “One specific case, Neural Collaborative Filtering (NCF) [15, 25] used as part of the MLPerf benchmark [19], uses an MLP rather than dot product to compute interactions between embeddings in matrix factorization.”
• Understanding of Machine Learning with Deep Learning: Architectures, Workflow, Applications and Future Directions (2023, methodology): “Sedhain et al. [72] proposed an AE-based collaborative filtering RS, referred to as AutoRec, where the original partial observed vectors have been replaced by integer ratings.”
• UvA-DARE ( Digital Academic Repository ) Graph Convolutional Matrix Completion (2017, positive): “We shall note at this point that several recent stateof-the-art models for matrix completion [17, 27, 7, 32] can be cast into this framework and understood as a special case of our model.”
• Disentangled Graph Collaborative Filtering (2020, background): “Another similar line applies autoencoders on interaction histories to estimate the generative process of user behaviors, such as Mult-VAE [23], AutoRec [31], and CDAE [44].”