One dimensional fractional order : Gamma-convergence and bilevel training scheme
arXiv:1612.05142
Abstract
New fractional -order seminorms, , , , are proposed in the one-dimensional (1D) setting, as a generalization of the integer order -seminorms, . The fractional -order -seminorms are shown to be intermediate between the integer order -seminorms. A bilevel training scheme is proposed, where under a box constraint a simultaneous optimization with respect to parameters and order of derivation is performed. Existence of solutions to the bilevel training scheme is proved by -convergence. Finally, the numerical landscape of the cost function associated to the bilevel training scheme is discussed for two numerical examples.
arXiv comment4 figures
References in corpus (5)
- A combined first and second order variational approach for image reconstruction
- Learning Graphical Model Parameters with Approximate Marginal Inference
- Insights into analysis operator learning: From patch-based sparse models to higher-order MRFs
- The structure of optimal parameters for image restoration problems
- Explorations on anisotropic regularisation of dynamic inverse problems by bilevel optimisation