Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.16277 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909968670654464 |
|---|---|
| author | Wang, Jialiang Bao, Xueyan Wu, Hao |
| author_facet | Wang, Jialiang Bao, Xueyan Wu, Hao |
| contents | Second-order Latent Factor (SLF) model, a class of low-rank representation learning methods, has proven effective at extracting node-to-node interaction patterns from High-dimensional and Incomplete (HDI) data. However, its optimization is notoriously difficult due to its bilinear and non-convex nature. Sharpness-aware Minimization (SAM) has recently proposed to find flat local minima when minimizing non-convex objectives, thereby improving the generalization of representation-learning models. To address this challenge, we propose a Sharpness-aware SLF (SSLF) model. SSLF embodies two key ideas: (1) acquiring second-order information via Hessian-vector products; and (2) injecting a sharpness term into the curvature (Hessian) through the designed Hessian-vector products. Experiments on multiple industrial datasets demonstrate that the proposed model consistently outperforms state-of-the-art baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_16277 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Sharpness-aware Second-order Latent Factor Model for High-dimensional and Incomplete Data Wang, Jialiang Bao, Xueyan Wu, Hao Machine Learning Second-order Latent Factor (SLF) model, a class of low-rank representation learning methods, has proven effective at extracting node-to-node interaction patterns from High-dimensional and Incomplete (HDI) data. However, its optimization is notoriously difficult due to its bilinear and non-convex nature. Sharpness-aware Minimization (SAM) has recently proposed to find flat local minima when minimizing non-convex objectives, thereby improving the generalization of representation-learning models. To address this challenge, we propose a Sharpness-aware SLF (SSLF) model. SSLF embodies two key ideas: (1) acquiring second-order information via Hessian-vector products; and (2) injecting a sharpness term into the curvature (Hessian) through the designed Hessian-vector products. Experiments on multiple industrial datasets demonstrate that the proposed model consistently outperforms state-of-the-art baselines. |
| title | Sharpness-aware Second-order Latent Factor Model for High-dimensional and Incomplete Data |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2512.16277 |