Power-law scaling to assist with key challenges in artificial intelligence

Power-law scaling, a central concept in critical phenomena, is found to be useful in deep learning, where optimized test errors on handwritten digit examples converge as a power-law to zero with database size. For rapid decision making with one training epoch, each example is presented only once to...

Descripción completa

Detalles Bibliográficos
Autores principales: Meir, Yuval, Sardi, Shira, Hodassman, Shiri, Kisos, Karin, Ben-Noam, Itamar, Goldental, Amir, Kanter, Ido
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Nature Publishing Group UK 2020
Materias:
Article
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7665018/
https://www.ncbi.nlm.nih.gov/pubmed/33184422
http://dx.doi.org/10.1038/s41598-020-76764-1

Internet

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7665018/
https://www.ncbi.nlm.nih.gov/pubmed/33184422
http://dx.doi.org/10.1038/s41598-020-76764-1

Ejemplares similares