Feature-Based Interpretation of the Deep Neural Network

The significant advantage of deep neural networks is that the upper layer can capture the high-level features of data based on the information acquired from the lower layer by stacking layers deeply. Since it is challenging to interpret what knowledge the neural network has learned, various studies...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Eun-Hun Lee, Hyeoncheol Kim
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	neural network explainable artificial intelligence (XAI) interpretability Electronics TK7800-8360
Acceso en línea:	https://doaj.org/article/8e911b6ca70f453995ea77a9bb616082
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:8e911b6ca70f453995ea77a9bb616082
record_format	dspace
spelling	oai:doaj.org-article:8e911b6ca70f453995ea77a9bb6160822021-11-11T15:40:43ZFeature-Based Interpretation of the Deep Neural Network10.3390/electronics102126872079-9292https://doaj.org/article/8e911b6ca70f453995ea77a9bb6160822021-11-01T00:00:00Zhttps://www.mdpi.com/2079-9292/10/21/2687https://doaj.org/toc/2079-9292The significant advantage of deep neural networks is that the upper layer can capture the high-level features of data based on the information acquired from the lower layer by stacking layers deeply. Since it is challenging to interpret what knowledge the neural network has learned, various studies for explaining neural networks have emerged to overcome this problem. However, these studies generate the local explanation of a single instance rather than providing a generalized global interpretation of the neural network model itself. To overcome such drawbacks of the previous approaches, we propose the global interpretation method for the deep neural network through features of the model. We first analyzed the relationship between the input and hidden layers to represent the high-level features of the model, then interpreted the decision-making process of neural networks through high-level features. In addition, we applied network pruning techniques to make concise explanations and analyzed the effect of layer complexity on interpretability. We present experiments on the proposed approach using three different datasets and show that our approach could generate global explanations on deep neural network models with high accuracy and fidelity.Eun-Hun LeeHyeoncheol KimMDPI AGarticleneural networkexplainable artificial intelligence (XAI)interpretabilityElectronicsTK7800-8360ENElectronics, Vol 10, Iss 2687, p 2687 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	neural network explainable artificial intelligence (XAI) interpretability Electronics TK7800-8360
spellingShingle	neural network explainable artificial intelligence (XAI) interpretability Electronics TK7800-8360 Eun-Hun Lee Hyeoncheol Kim Feature-Based Interpretation of the Deep Neural Network
description	The significant advantage of deep neural networks is that the upper layer can capture the high-level features of data based on the information acquired from the lower layer by stacking layers deeply. Since it is challenging to interpret what knowledge the neural network has learned, various studies for explaining neural networks have emerged to overcome this problem. However, these studies generate the local explanation of a single instance rather than providing a generalized global interpretation of the neural network model itself. To overcome such drawbacks of the previous approaches, we propose the global interpretation method for the deep neural network through features of the model. We first analyzed the relationship between the input and hidden layers to represent the high-level features of the model, then interpreted the decision-making process of neural networks through high-level features. In addition, we applied network pruning techniques to make concise explanations and analyzed the effect of layer complexity on interpretability. We present experiments on the proposed approach using three different datasets and show that our approach could generate global explanations on deep neural network models with high accuracy and fidelity.
format	article
author	Eun-Hun Lee Hyeoncheol Kim
author_facet	Eun-Hun Lee Hyeoncheol Kim
author_sort	Eun-Hun Lee
title	Feature-Based Interpretation of the Deep Neural Network
title_short	Feature-Based Interpretation of the Deep Neural Network
title_full	Feature-Based Interpretation of the Deep Neural Network
title_fullStr	Feature-Based Interpretation of the Deep Neural Network
title_full_unstemmed	Feature-Based Interpretation of the Deep Neural Network
title_sort	feature-based interpretation of the deep neural network
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/8e911b6ca70f453995ea77a9bb616082
work_keys_str_mv	AT eunhunlee featurebasedinterpretationofthedeepneuralnetwork AT hyeoncheolkim featurebasedinterpretationofthedeepneuralnetwork
_version_	1718434424029184000

Feature-Based Interpretation of the Deep Neural Network

Ejemplares similares