Cargando…

Machine learning in spectral domain

Deep neural networks are usually trained in the space of the nodes, by adjusting the weights of existing links via suitable optimization protocols. We here propose a radically new approach which anchors the learning process to reciprocal space. Specifically, the training acts on the spectral domain...

Descripción completa

Detalles Bibliográficos
Autores principales: Giambagli, Lorenzo, Buffoni, Lorenzo, Carletti, Timoteo, Nocentini, Walter, Fanelli, Duccio
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Nature Publishing Group UK 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7910623/
https://www.ncbi.nlm.nih.gov/pubmed/33637729
http://dx.doi.org/10.1038/s41467-021-21481-0

Ejemplares similares