Collaborative Recurrent Autoencoder: Recommend while Learning to Fill in the Blanks

By Hao Wang, Xingjian Shi, Dit-Yan Yeung. 2016

107 citations recorded by the source corpus.

Hybrid methods that utilize both content and rating information are commonly used in many recommender systems. However, most of them use either handcrafted features or the bag-of-words representation as a surrogate for the content information but they are neither effective nor natural enough. To address this problem, we develop a collaborative recurrent autoencoder (CRAE) which is a denoising recurrent autoencoder (DRAE) that models the generation of content sequences in the collaborative filtering (CF) setting. The model generalizes recent advances in recurrent deep learning from i.i.d. input to non-i.i.d. (CF-based) input and provides a new denoising scheme along with a novel learnable pooling scheme for the recurrent autoencoder. To do this, we first develop a hierarchical Bayesian model for the DRAE and then generalize it to the CF setting. The synergy between denoising and CF enables CRAE to make accurate recommendations while learning to fill in the blanks in sequences. Experiments on real-world datasets from different domains (CiteULike and Netflix) show that, by jointly modeling the order-aware generation of sequences for the content information and performing CF for the ratings, CRAE is able to significantly outperform the state of the art on both the recommendation task based on ratings and the sequence generation task based on content information.

Subfields: Computer Vision and Pattern Recognition, Information 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. Xiang Wang
3. Yi Tay
4. Meng Wang
5. Hao Wang
6. Dit-Yan Yeung
7. Le Wu
8. Kun Zhang

In Information Systems

1. Xiangnan He
2. Xiang Wang
3. Meng Wang
4. Yi Tay
5. Le Wu
6. Dit-Yan Yeung
7. Hao Wang
8. Kun Zhang

In Signal Processing

1. Dit-Yan Yeung
2. Hao Wang
3. Yi Tay
4. Ali Yurekli
5. Cihan Kaleli
6. Alper Bilge
7. Zeynep Batmaz
8. Xiangnan He

Citations in the corpus

• Knowledge-aware Autoencoders for Explainable Recommender Systems (2018, background): “More recently, in [21] the authors propose a hybrid architecture for Autoencoders in order to incorporate both users’ feedbacks and content description about items.”
• A review on deep learning for recommender systems: challenges and remedies (2018, background): “Wang et al. (2016) propose a collaborative recurrent autoencoder to generate a hybrid recommender system to improve recommendation accuracy by jointly modeling generation of sequences and implicit relationships between users and items.”
• A Survey on Bayesian Deep Learning (2016, methodology): “It is also worth noting that the perception component can be a recurrent version of probabilistic SDAE [122] or VAE [17, 68] to handle raw sequential data, while the task-specific component can take more sophisticated forms to accommodate more complex recommendation scenarios (e.”
• Deep Learning Based Recommender System (2017, methodology): “Sequential Information w/t User ID [16, 29, 33, 35, 73, 91, 118, 134, 144, 161, 174, 176, 190, 195, 199, 206]”
• Cross-domain Recommendation via Deep Domain Adaptation (2018, methodology): “ reason, the application of deep learning to recommender systems has gained research a−ention. Wang et al. [37] and Li et al. [23] combined autoencoder models with collaborative •ltering; Wang et al. [36] used a RNN model for item representation and combined it with CF; Wu et al. [38] modeled tempor”
• A Survey on Accuracy-Oriented Neural Recommendation: From Collaborative Filtering to Information-Rich Recommendation (2021, background): “the bottleneck hidden content representations of items [30], [54], [71], [72], [73], [74], [75], [76], [77], [78].”
• Collaborative Variational Autoencoder for Recommender Systems (2017, background): “Very recently, collaborative deep learning (CDL) [29] and collaborative recurrent autoencoder [30] have been proposed for joint learning a stacked denoising autoencoder (SDAE) [26] (or denoising recurrent autoencoder) and collaborative €ltering, and they shows promising performance.”
• When large language models meet personalization: perspectives of challenges and opportunities (2023, unknown): “CRAE [87] introduces a collaborative recurrent autoencoder that captures the word order in texts, enabling the modeling of content sequences in collaborative filtering scenarios.”