Learning to Learn Weight Generation via Local Consistency Diffusion
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909614001356800 |
|---|---|
| author | Guan, Yunchuan Liu, Yu Zhou, Ke Shen, Zhiqi Hwang, Jenq-Neng Li, Lei |
| author_facet | Guan, Yunchuan Liu, Yu Zhou, Ke Shen, Zhiqi Hwang, Jenq-Neng Li, Lei |
| contents | Diffusion-based algorithms have emerged as promising techniques for weight generation. However, existing solutions are limited by two challenges: generalizability and local target assignment. The former arises from the inherent lack of cross-task transferability in existing single-level optimization methods, limiting the model's performance on new tasks. The latter lies in existing research modeling only global optimal weights, neglecting the supervision signals in local target weights. Moreover, naively assigning local target weights causes local-global inconsistency. To address these issues, we propose Mc-Di, which integrates the diffusion algorithm with meta-learning for better generalizability. Furthermore, we extend the vanilla diffusion into a local consistency diffusion algorithm. Our theory and experiments demonstrate that it can learn from local targets while maintaining consistency with the global optima. We validate Mc-Di's superior accuracy and inference efficiency in tasks that require frequent weight updates, including transfer learning, few-shot learning, domain generalization, and large language model adaptation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_01117 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Learning to Learn Weight Generation via Local Consistency Diffusion Guan, Yunchuan Liu, Yu Zhou, Ke Shen, Zhiqi Hwang, Jenq-Neng Li, Lei Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition Diffusion-based algorithms have emerged as promising techniques for weight generation. However, existing solutions are limited by two challenges: generalizability and local target assignment. The former arises from the inherent lack of cross-task transferability in existing single-level optimization methods, limiting the model's performance on new tasks. The latter lies in existing research modeling only global optimal weights, neglecting the supervision signals in local target weights. Moreover, naively assigning local target weights causes local-global inconsistency. To address these issues, we propose Mc-Di, which integrates the diffusion algorithm with meta-learning for better generalizability. Furthermore, we extend the vanilla diffusion into a local consistency diffusion algorithm. Our theory and experiments demonstrate that it can learn from local targets while maintaining consistency with the global optima. We validate Mc-Di's superior accuracy and inference efficiency in tasks that require frequent weight updates, including transfer learning, few-shot learning, domain generalization, and large language model adaptation. |
| title | Learning to Learn Weight Generation via Local Consistency Diffusion |
| topic | Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2502.01117 |