Zobrazeno 1 - 3
of 3
pro vyhledávání: '"Keegan Stoner"'
Publikováno v:
Machine Learning: Science and Technology, Vol 5, Iss 1, p 015002 (2024)
Both the path integral measure in field theory (FT) and ensembles of neural networks (NN) describe distributions over functions. When the central limit theorem can be applied in the infinite-width (infinite- N ) limit, the ensemble of networks corres
Externí odkaz:
https://doaj.org/article/03950f1b50da4580bcfd77385e25b245
We propose a theoretical understanding of neural networks in terms of Wilsonian effective field theory. The correspondence relies on the fact that many asymptotic neural networks are drawn from Gaussian processes, the analog of non-interacting field
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::9f4281635f4b8bdb9d463923a26e7ef5
