Home

Published

- 2 min read

A new perspective on optimizers: leveraging moreau-yosida approx. in learning

NeuralNetworks Foundations
img of A new perspective on optimizers: leveraging moreau-yosida approx. in learning

Journal Paper

Details

The largely known notion of optimizer is revisited in a more general fashion, under the lens of Moreau-Yosida approximation, leading to novel ideas in the context of learning over time.

  • Authors: Stefano Melacci, Alessandro Betti, Michele Casoni, Tommaso Guidi, Matteo Tiezzi, Marco Gori
  • Title: A new perspective on optimizers: leveraging moreau-yosida approximation in gradient-based learning
  • Where: Intelligenza Artificiale (iOS Press) 2024

BibTeX

   @article{doi:10.3233/IA-240047,
	author = {Alessandro Betti and Gabriele Ciravegna and Marco Gori and Stefano Melacci and Kevin Mottin and Frédéric Precioso},
	title ={A new perspective on optimizers: leveraging moreau-yosida approximation in gradient-based learning},
	journal = {Intelligenza Artificiale},
	volume = {18},
	number = {2},
	pages = {301-311},
	year = {2024},
	doi = {10.3233/IA-240047}
}