Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.21541 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917357681639424 |
|---|---|
| author | Li, Yawen Hu, Tao Lian, Zhouhui Tian, Wan Peng, Yijie Zhang, Huiming Li, Zhongyi |
| author_facet | Li, Yawen Hu, Tao Lian, Zhouhui Tian, Wan Peng, Yijie Zhang, Huiming Li, Zhongyi |
| contents | This paper studies generalization error bounds for Transformer models. Based on the offset Rademacher complexity, we derive sharper generalization bounds for different Transformer architectures, including single-layer single-head, single-layer multi-head, and multi-layer Transformers. We first express the excess risk of Transformers in terms of the offset Rademacher complexity. By exploiting its connection with the empirical covering numbers of the corresponding hypothesis spaces, we obtain excess risk bounds that achieve optimal convergence rates up to constant factors. We then derive refined excess risk bounds by upper bounding the covering numbers of Transformer hypothesis spaces using matrix ranks and matrix norms, leading to precise, architecture-dependent generalization bounds. Finally, we relax the boundedness assumption on feature mappings and extend our theoretical results to settings with unbounded (sub-Gaussian) features and heavy-tailed distributions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_21541 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Sharper Generalization Bounds for Transformer Li, Yawen Hu, Tao Lian, Zhouhui Tian, Wan Peng, Yijie Zhang, Huiming Li, Zhongyi Machine Learning Artificial Intelligence This paper studies generalization error bounds for Transformer models. Based on the offset Rademacher complexity, we derive sharper generalization bounds for different Transformer architectures, including single-layer single-head, single-layer multi-head, and multi-layer Transformers. We first express the excess risk of Transformers in terms of the offset Rademacher complexity. By exploiting its connection with the empirical covering numbers of the corresponding hypothesis spaces, we obtain excess risk bounds that achieve optimal convergence rates up to constant factors. We then derive refined excess risk bounds by upper bounding the covering numbers of Transformer hypothesis spaces using matrix ranks and matrix norms, leading to precise, architecture-dependent generalization bounds. Finally, we relax the boundedness assumption on feature mappings and extend our theoretical results to settings with unbounded (sub-Gaussian) features and heavy-tailed distributions. |
| title | Sharper Generalization Bounds for Transformer |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2603.21541 |