Using phidelta diagrams to discover relevant patterns in multilayer perceptrons

Abstract Understanding the inner behaviour of multilayer perceptrons during and after training is a goal of paramount importance for many researchers worldwide. This article experimentally shows that relevant patterns emerge upon training, which are typically related to the underlying problem diffic...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autor principal:	Giuliano Armano
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2020
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/23b3497c05374c0a8f955325aa4f0c7d
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:23b3497c05374c0a8f955325aa4f0c7d
record_format	dspace
spelling	oai:doaj.org-article:23b3497c05374c0a8f955325aa4f0c7d2021-12-02T11:40:20ZUsing phidelta diagrams to discover relevant patterns in multilayer perceptrons10.1038/s41598-020-76517-02045-2322https://doaj.org/article/23b3497c05374c0a8f955325aa4f0c7d2020-12-01T00:00:00Zhttps://doi.org/10.1038/s41598-020-76517-0https://doaj.org/toc/2045-2322Abstract Understanding the inner behaviour of multilayer perceptrons during and after training is a goal of paramount importance for many researchers worldwide. This article experimentally shows that relevant patterns emerge upon training, which are typically related to the underlying problem difficulty. The occurrence of these patterns is highlighted by means of $$\langle \varphi ,\delta \rangle$$ ⟨ φ , δ ⟩ diagrams, a 2D graphical tool originally devised to support the work of researchers on classifier performance evaluation and on feature assessment. The underlying assumption being that multilayer perceptrons are powerful engines for feature encoding, hidden layers have been inspected as they were in fact hosting new input features. Interestingly, there are problems that appear difficult if dealt with using a single hidden layer, whereas they turn out to be easier upon the addition of further layers. The experimental findings reported in this article give further support to the standpoint according to which implementing neural architectures with multiple layers may help to boost their generalisation ability. A generic training strategy inspired by some relevant recommendations of deep learning has also been devised. A basic implementation of this strategy has been thoroughly used during the experiments aimed at identifying relevant patterns inside multilayer perceptrons. Further experiments performed in a comparative setting have shown that it could be adopted as viable alternative to the classical backpropagation algorithm.Giuliano ArmanoNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 10, Iss 1, Pp 1-17 (2020)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Giuliano Armano Using phidelta diagrams to discover relevant patterns in multilayer perceptrons
description	Abstract Understanding the inner behaviour of multilayer perceptrons during and after training is a goal of paramount importance for many researchers worldwide. This article experimentally shows that relevant patterns emerge upon training, which are typically related to the underlying problem difficulty. The occurrence of these patterns is highlighted by means of $$\langle \varphi ,\delta \rangle$$ ⟨ φ , δ ⟩ diagrams, a 2D graphical tool originally devised to support the work of researchers on classifier performance evaluation and on feature assessment. The underlying assumption being that multilayer perceptrons are powerful engines for feature encoding, hidden layers have been inspected as they were in fact hosting new input features. Interestingly, there are problems that appear difficult if dealt with using a single hidden layer, whereas they turn out to be easier upon the addition of further layers. The experimental findings reported in this article give further support to the standpoint according to which implementing neural architectures with multiple layers may help to boost their generalisation ability. A generic training strategy inspired by some relevant recommendations of deep learning has also been devised. A basic implementation of this strategy has been thoroughly used during the experiments aimed at identifying relevant patterns inside multilayer perceptrons. Further experiments performed in a comparative setting have shown that it could be adopted as viable alternative to the classical backpropagation algorithm.
format	article
author	Giuliano Armano
author_facet	Giuliano Armano
author_sort	Giuliano Armano
title	Using phidelta diagrams to discover relevant patterns in multilayer perceptrons
title_short	Using phidelta diagrams to discover relevant patterns in multilayer perceptrons
title_full	Using phidelta diagrams to discover relevant patterns in multilayer perceptrons
title_fullStr	Using phidelta diagrams to discover relevant patterns in multilayer perceptrons
title_full_unstemmed	Using phidelta diagrams to discover relevant patterns in multilayer perceptrons
title_sort	using phidelta diagrams to discover relevant patterns in multilayer perceptrons
publisher	Nature Portfolio
publishDate	2020
url	https://doaj.org/article/23b3497c05374c0a8f955325aa4f0c7d
work_keys_str_mv	AT giulianoarmano usingphideltadiagramstodiscoverrelevantpatternsinmultilayerperceptrons
_version_	1718395630321139712

Using phidelta diagrams to discover relevant patterns in multilayer perceptrons

Ejemplares similares