S-Diff: An Anisotropic Diffusion Model for Collaborative Filtering in Spectral Domain
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_ | 1866909446064570368 |
|---|---|
| author | Xia, Rui Cheng, Yanhua Tang, Yongxiang Liu, Xiaocheng Liu, Xialong Wang, Lisong Jiang, Peng |
| author_facet | Xia, Rui Cheng, Yanhua Tang, Yongxiang Liu, Xiaocheng Liu, Xialong Wang, Lisong Jiang, Peng |
| contents | Recovering user preferences from user-item interaction matrices is a key challenge in recommender systems. While diffusion models can sample and reconstruct preferences from latent distributions, they often fail to capture similar users' collective preferences effectively. Additionally, latent variables degrade into pure Gaussian noise during the forward process, lowering the signal-to-noise ratio, which in turn degrades performance. To address this, we propose S-Diff, inspired by graph-based collaborative filtering, better to utilize low-frequency components in the graph spectral domain. S-Diff maps user interaction vectors into the spectral domain and parameterizes diffusion noise to align with graph frequency. This anisotropic diffusion retains significant low-frequency components, preserving a high signal-to-noise ratio. S-Diff further employs a conditional denoising network to encode user interactions, recovering true preferences from noisy data. This method achieves strong results across multiple datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_00384 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | S-Diff: An Anisotropic Diffusion Model for Collaborative Filtering in Spectral Domain Xia, Rui Cheng, Yanhua Tang, Yongxiang Liu, Xiaocheng Liu, Xialong Wang, Lisong Jiang, Peng Information Retrieval Recovering user preferences from user-item interaction matrices is a key challenge in recommender systems. While diffusion models can sample and reconstruct preferences from latent distributions, they often fail to capture similar users' collective preferences effectively. Additionally, latent variables degrade into pure Gaussian noise during the forward process, lowering the signal-to-noise ratio, which in turn degrades performance. To address this, we propose S-Diff, inspired by graph-based collaborative filtering, better to utilize low-frequency components in the graph spectral domain. S-Diff maps user interaction vectors into the spectral domain and parameterizes diffusion noise to align with graph frequency. This anisotropic diffusion retains significant low-frequency components, preserving a high signal-to-noise ratio. S-Diff further employs a conditional denoising network to encode user interactions, recovering true preferences from noisy data. This method achieves strong results across multiple datasets. |
| title | S-Diff: An Anisotropic Diffusion Model for Collaborative Filtering in Spectral Domain |
| topic | Information Retrieval |
| url | https://arxiv.org/abs/2501.00384 |