Neuronal diversity can improve machine learning for physics and beyond

Diversity conveys advantages in nature, yet homogeneous neurons typically comprise the layers of artificial neural networks. Here we construct neural networks from neurons that learn their own activation functions, quickly diversify, and subsequently outperform their homogeneous counterparts on imag...

Descripción completa

Detalles Bibliográficos
Autores principales: Choudhary, Anshul, Radhakrishnan, Anil, Lindner, John F., Sinha, Sudeshna, Ditto, William L.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Nature Publishing Group UK 2023
Materias:
Article
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10460398/
https://www.ncbi.nlm.nih.gov/pubmed/37634029
http://dx.doi.org/10.1038/s41598-023-40766-6

Internet

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10460398/
https://www.ncbi.nlm.nih.gov/pubmed/37634029
http://dx.doi.org/10.1038/s41598-023-40766-6

Ejemplares similares