Federica Porta
University of Ferrara
35 Papers
28 Citations
Federica Porta is an academic researcher from University of Ferrara. The author has contributed to research in topics: Computer science & Proximal Gradient Methods. The author has an hindex of 9, co-authored 23 publications. Previous affiliations of Federica Porta include University of Modena and Reggio Emilia.
Chat about Author
Papers
Variable metric inexact line-search based methods for nonsmooth optimization
TL;DR: In this article, a proximal gradient method for minimizing the sum of a differentiable, possibly nonconvex, function plus a convex, possibly nondifferentiable, function is proposed.
109
A New Steplength Selection for Scaled Gradient Methods with Application to Image Deblurring
TL;DR: This paper extends the rule for the steplength selection approximating the inverse of some eigenvalues of the Hessian matrix to the case of scaled gradient projection methods applied to constrained minimization problems, and tests the effectiveness of the proposed strategy in image deblurring problems.
39
Neural architecture search via standard machine learning methodologies
TL;DR: By a probabilistic exploration of the hyperparameter space, the main contribution of the paper consists in introducing an automatic Machine Learning technique to set these hyperparameters in such a way that a measure of the CNN performance can be optimised.
23
A Line Search Based Proximal Stochastic Gradient Algorithm with Dynamical Variance Reduction
TL;DR: A proper technique to dynamically reduce the variance of the stochastic gradients along the iterative process with a descent condition in expectation for the objective function, aimed to set the value for the steplength parameter at each iteration is developed.
13
Serial and parallel approaches for image segmentation by numerical minimization of a second-order functional
TL;DR: It is proved that this parallel method (OPARBCDA) generates a sequence of iterates which converges to a critical point of the functional on the level set devised by the starting point and it is shown that the parallel method can be efficiently implemented even in a commodity multicore CPU.
10