Predicting trends in the quality of state-of-the-art neural networks without access to training or testing data

In many machine learning applications, one uses pre-trained neural networks, having limited access to training and test data. Martin et al. show how to predict trends in the quality of such neural networks without access to this information, relevant for reproducibility, diagnostics, and validation.

Guardado en:

Detalles Bibliográficos
Autores principales:	Charles H. Martin, Tongsu (Serena) Peng, Michael W. Mahoney
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Science Q
Acceso en línea:	https://doaj.org/article/bdc9afbd811d47888c4645cf78e0b595
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:bdc9afbd811d47888c4645cf78e0b595
record_format	dspace
spelling	oai:doaj.org-article:bdc9afbd811d47888c4645cf78e0b5952021-12-02T18:34:21ZPredicting trends in the quality of state-of-the-art neural networks without access to training or testing data10.1038/s41467-021-24025-82041-1723https://doaj.org/article/bdc9afbd811d47888c4645cf78e0b5952021-07-01T00:00:00Zhttps://doi.org/10.1038/s41467-021-24025-8https://doaj.org/toc/2041-1723In many machine learning applications, one uses pre-trained neural networks, having limited access to training and test data. Martin et al. show how to predict trends in the quality of such neural networks without access to this information, relevant for reproducibility, diagnostics, and validation.Charles H. MartinTongsu (Serena) PengMichael W. MahoneyNature PortfolioarticleScienceQENNature Communications, Vol 12, Iss 1, Pp 1-13 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Science Q
spellingShingle	Science Q Charles H. Martin Tongsu (Serena) Peng Michael W. Mahoney Predicting trends in the quality of state-of-the-art neural networks without access to training or testing data
description	In many machine learning applications, one uses pre-trained neural networks, having limited access to training and test data. Martin et al. show how to predict trends in the quality of such neural networks without access to this information, relevant for reproducibility, diagnostics, and validation.
format	article
author	Charles H. Martin Tongsu (Serena) Peng Michael W. Mahoney
author_facet	Charles H. Martin Tongsu (Serena) Peng Michael W. Mahoney
author_sort	Charles H. Martin
title	Predicting trends in the quality of state-of-the-art neural networks without access to training or testing data
title_short	Predicting trends in the quality of state-of-the-art neural networks without access to training or testing data
title_full	Predicting trends in the quality of state-of-the-art neural networks without access to training or testing data
title_fullStr	Predicting trends in the quality of state-of-the-art neural networks without access to training or testing data
title_full_unstemmed	Predicting trends in the quality of state-of-the-art neural networks without access to training or testing data
title_sort	predicting trends in the quality of state-of-the-art neural networks without access to training or testing data
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/bdc9afbd811d47888c4645cf78e0b595
work_keys_str_mv	AT charleshmartin predictingtrendsinthequalityofstateoftheartneuralnetworkswithoutaccesstotrainingortestingdata AT tongsuserenapeng predictingtrendsinthequalityofstateoftheartneuralnetworkswithoutaccesstotrainingortestingdata AT michaelwmahoney predictingtrendsinthequalityofstateoftheartneuralnetworkswithoutaccesstotrainingortestingdata
_version_	1718377884082503680

Predicting trends in the quality of state-of-the-art neural networks without access to training or testing data

Ejemplares similares