Mamba4Rec: Towards Efficient Sequential Recommendation with Selective State Space Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929403201585152 |
|---|---|
| author | Liu, Chengkai Lin, Jianghao Wang, Jianling Liu, Hanzhou Caverlee, James |
| author_facet | Liu, Chengkai Lin, Jianghao Wang, Jianling Liu, Hanzhou Caverlee, James |
| contents | Sequential recommendation aims to estimate the dynamic user preferences and sequential dependencies among historical user behaviors. Although Transformer-based models have proven to be effective for sequential recommendation, they suffer from the inference inefficiency problem stemming from the quadratic computational complexity of attention operators, especially for long behavior sequences. Inspired by the recent success of state space models (SSMs), we propose Mamba4Rec, which is the first work to explore the potential of selective SSMs for efficient sequential recommendation. Built upon the basic Mamba block which is a selective SSM with an efficient hardware-aware parallel algorithm, we design a series of sequential modeling techniques to further promote model performance while maintaining inference efficiency. Through experiments on public datasets, we demonstrate how Mamba4Rec effectively tackles the effectiveness-efficiency dilemma, outperforming both RNN- and attention-based baselines in terms of both effectiveness and efficiency. The code is available at https://github.com/chengkai-liu/Mamba4Rec. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_03900 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Mamba4Rec: Towards Efficient Sequential Recommendation with Selective State Space Models Liu, Chengkai Lin, Jianghao Wang, Jianling Liu, Hanzhou Caverlee, James Information Retrieval Sequential recommendation aims to estimate the dynamic user preferences and sequential dependencies among historical user behaviors. Although Transformer-based models have proven to be effective for sequential recommendation, they suffer from the inference inefficiency problem stemming from the quadratic computational complexity of attention operators, especially for long behavior sequences. Inspired by the recent success of state space models (SSMs), we propose Mamba4Rec, which is the first work to explore the potential of selective SSMs for efficient sequential recommendation. Built upon the basic Mamba block which is a selective SSM with an efficient hardware-aware parallel algorithm, we design a series of sequential modeling techniques to further promote model performance while maintaining inference efficiency. Through experiments on public datasets, we demonstrate how Mamba4Rec effectively tackles the effectiveness-efficiency dilemma, outperforming both RNN- and attention-based baselines in terms of both effectiveness and efficiency. The code is available at https://github.com/chengkai-liu/Mamba4Rec. |
| title | Mamba4Rec: Towards Efficient Sequential Recommendation with Selective State Space Models |
| topic | Information Retrieval |
| url | https://arxiv.org/abs/2403.03900 |