Self-supervised Learning for Large-scale Item Recommendations

By Tiansheng Yao, Xinyang Yi, Derek Zhiyuan Cheng, Felix X. Yu, Ting-Li Chen, Aditya Krishna Menon, Lichan Hong, Ed H. Chi, S. Tjoa, Jieqi Kang and 1 more. International Conference on Information and Knowledge Management, 2020

329 citations recorded by the source corpus.

Large scale recommender models find most relevant items from huge catalogs, and they play a critical role in modern search and recommendation systems. To model the input space with large-vocab categorical features, a typical recommender model learns a joint embedding space through neural networks for both queries and items from user feedback data. However, with millions to billions of items in the corpus, users tend to provide feedback for a very small set of them, causing a power-law distribution. This makes the feedback data for long-tail items extremely sparse. Inspired by the recent success in self-supervised representation learning research in both computer vision and natural language understanding, we propose a multi-task self-supervised learning (SSL) framework for large-scale item recommendations. The framework is designed to tackle the label sparsity problem by learning better latent relationship of item features. Specifically, SSL improves item representation learning as well as serving as additional regularization to improve generalization. Furthermore, we propose a novel data augmentation method that utilizes feature correlations within the proposed framework. We evaluate our framework using two real-world datasets with 500M and 1B training examples respectively. Our results demonstrate the effectiveness of SSL regularization and show its superior performance over the state-of-the-art regularization techniques. We also have already launched the proposed techniques to a web-scale commercial app-to-app recommendation system, with significant improvements top-tier business metrics demonstrated in A/B experiments on live traffic. Our online results also verify our hypothesis that our framework indeed improves model performance even more on slices that lack supervision.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Information Systems

View Paper: https://www.semanticscholar.org/paper/e54aa7f89746749e5f387f0d5d234ffd23c16cf8

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. Hongzhi Yin
2. Tong Chen
3. Xin Xia
4. Junliang Yu
5. Zi-Liang Huang
6. Julian McAuley
7. Li-zhen Cui
8. Chao Huang

In Computer Vision and Pattern Recognition

1. Zi-Liang Huang
2. Hongzhi Yin
3. Julian McAuley
4. Caiming Xiong
5. Junliang Yu
6. Tong Chen
7. Xin Xia
8. Ruihong Qiu

In Information Systems

1. Hongzhi Yin
2. Tong Chen
3. Junliang Yu
4. Xin Xia
5. Zi-Liang Huang
6. Li-zhen Cui
7. Chao Huang
8. Lianghao Xia

Citations in the corpus

• XSimGCL: Towards Extremely Simple Graph Contrastive Learning for Recommendation (2022, methodology): “Actually, this is a universal problem for all the CL-based recommendation models [12], [17], [25], [26], [27] following the paradigm in Fig.”
• Self-Supervised Learning for Recommender Systems: A Survey (2022, background): “SL4Rec [72] adopts a two-tower architecture and applies correlated feature masking and dropout on the item features for more meaningful feature augmentations.”
• Intent Contrastive Learning for Sequential Recommendation (2022, methodology): “Contrastive Self-Supervised Learning (SSL) has brought much attentions by different research communities including CV [2, 4, 9, 14, 20] and NLP [7, 8, 29, 50], as well as recommendation [44, 45, 47, 51].”
• Contrastive Learning for Representation Degeneration Problem in Sequential Recommendation (2021, background): “SSL [54] proposes a siamese network to encode the items as pre-training with embedding-level augmentations.”
• Hypergraph Contrastive Collaborative Filtering (2022, methodology): “Recall@20 0.0093 0.0142 0.0131 0.0103 0.0213 0.0222 0.0192 0.0319 0.0317 0.0211 0.0302 0.0280 0.0296 0.0285 0.0327 0.0344 2 × 10 − 4 NDCG@20 0.0049 0.0085 0.0099 0.0074 0.0158 0.0160 0.0144 0.0236 0.0243 0.0154 0.0225 0.0202 0.0219 0.0238 0.0249 0.0258 5 × 10 − 3 Recall@40 0.0196 0.0223 0.0202 0.021”
• Are Graph Augmentations Necessary?: Simple Graph Contrastive Learning for Recommendation (2021, methodology): “• DNN+SSL [35] is a recent DNN-based recommendation method which adopts the similar architecture in Fig.”