Vector Quantization for Recommender Systems: A Review and Outlook

Qijiong LIU^*, Xiaoyu DONG^*, Jiaren XIAO, Nuo CHEN, Hengchang HU, Jieming ZHU, Chenxu ZHU, Tetsuya SAKAI, and Xiao-Ming WU^#
*Equal contribution (co-first authors).
[Paper]

Abstract

Vector quantization, renowned for its unparalleled feature compression capabilities, has been a prominent topic in signal processing and machine learning research for several decades and remains widely utilized today. With the emergence of large models and generative AI, vector quantization has gained popularity in recommender systems, establishing itself as a preferred solution. This paper starts with a comprehensive review of vector quantization techniques, categorizing them into three types. Then it delves into a systematic taxonomy of vector quantization techniques for recommender systems, examining their applications from multiple perspectives. Furthermore, we thoroughly introduce research efforts made in diverse recommendation scenarios, including efficiency-oriented approaches and quality-oriented approaches. Finally, the survey analyzes the remaining challenges and anticipates future trends in the field, including the challenges associated with the training of vector quantization, the opportunities presented by large language models, and emerging trends in multimodal recommender systems. We hope this survey can pave the way for future researchers in the recommendation community and accelerate their exploration in this promising field.

Citation

@misc{liu2024vector,
      title={Vector Quantization for Recommender Systems: A Review and Outlook}, 
      author={Qijiong Liu and Xiaoyu Dong and Jiaren Xiao and Nuo Chen and Hengchang Hu and Jieming Zhu and Chenxu Zhu and Tetsuya Sakai and Xiao-Ming Wu},
      year={2024},
      archivePrefix={arXiv},
      primaryClass={cs.IR}
}