Descent-Net: Learning Descent Directions for Constrained Optimization
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_ | 1866915674583990272 |
|---|---|
| author | Zhou, Zisheng Zheng, Dengyu Chen, Zirui Chen, Shixiang |
| author_facet | Zhou, Zisheng Zheng, Dengyu Chen, Zirui Chen, Shixiang |
| contents | Deep learning approaches, known for their ability to model complex relationships and fast execution, are increasingly being applied to solve large optimization problems. However, existing methods often face challenges in simultaneously ensuring feasibility and achieving an optimal objective value. To address this issue, we propose Descent-Net, a neural network designed to learn an effective descent direction from a feasible solution. By updating the solution along this learned direction, Descent-Net improves the objective value while preserving feasibility. Our method demonstrates strong performance on both synthetic optimization tasks and the real-world AC optimal power flow problem, while also exhibiting effective scalability to large problems, as shown by portfolio optimization experiments with thousands of assets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_11396 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Descent-Net: Learning Descent Directions for Constrained Optimization Zhou, Zisheng Zheng, Dengyu Chen, Zirui Chen, Shixiang Optimization and Control Deep learning approaches, known for their ability to model complex relationships and fast execution, are increasingly being applied to solve large optimization problems. However, existing methods often face challenges in simultaneously ensuring feasibility and achieving an optimal objective value. To address this issue, we propose Descent-Net, a neural network designed to learn an effective descent direction from a feasible solution. By updating the solution along this learned direction, Descent-Net improves the objective value while preserving feasibility. Our method demonstrates strong performance on both synthetic optimization tasks and the real-world AC optimal power flow problem, while also exhibiting effective scalability to large problems, as shown by portfolio optimization experiments with thousands of assets. |
| title | Descent-Net: Learning Descent Directions for Constrained Optimization |
| topic | Optimization and Control |
| url | https://arxiv.org/abs/2512.11396 |