A primal-dual algorithm for image reconstruction with input-convex neural network regularizers
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866917034471718912 |
|---|---|
| author | Ehrhardt, Matthias J. Mukherjee, Subhadip Wong, Hok Shing |
| author_facet | Ehrhardt, Matthias J. Mukherjee, Subhadip Wong, Hok Shing |
| contents | We address the optimization problem in a data-driven variational reconstruction framework, where the regularizer is parameterized by an input-convex neural network (ICNN). While gradient-based methods are commonly used to solve such problems, they struggle to effectively handle non-smooth problems which often leads to slow convergence. Moreover, the nested structure of the neural network complicates the application of standard non-smooth optimization techniques, such as proximal algorithms. To overcome these challenges, we reformulate the problem and eliminate the network's nested structure. By relating this reformulation to epigraphical projections of the activation functions, we transform the problem into a convex optimization problem that can be efficiently solved using a primal-dual algorithm. We also prove that this reformulation is equivalent to the original variational problem. Through experiments on several imaging tasks, we show that the proposed approach not only outperforms subgradient methods and even accelerated methods in the smooth setting, but also facilitates the training of the regularizer itself. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_12441 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | A primal-dual algorithm for image reconstruction with input-convex neural network regularizers Ehrhardt, Matthias J. Mukherjee, Subhadip Wong, Hok Shing Optimization and Control Computer Vision and Pattern Recognition Numerical Analysis 65K10, 90C25, 94A08 We address the optimization problem in a data-driven variational reconstruction framework, where the regularizer is parameterized by an input-convex neural network (ICNN). While gradient-based methods are commonly used to solve such problems, they struggle to effectively handle non-smooth problems which often leads to slow convergence. Moreover, the nested structure of the neural network complicates the application of standard non-smooth optimization techniques, such as proximal algorithms. To overcome these challenges, we reformulate the problem and eliminate the network's nested structure. By relating this reformulation to epigraphical projections of the activation functions, we transform the problem into a convex optimization problem that can be efficiently solved using a primal-dual algorithm. We also prove that this reformulation is equivalent to the original variational problem. Through experiments on several imaging tasks, we show that the proposed approach not only outperforms subgradient methods and even accelerated methods in the smooth setting, but also facilitates the training of the regularizer itself. |
| title | A primal-dual algorithm for image reconstruction with input-convex neural network regularizers |
| topic | Optimization and Control Computer Vision and Pattern Recognition Numerical Analysis 65K10, 90C25, 94A08 |
| url | https://arxiv.org/abs/2410.12441 |