Internal and collective interpretation for improving human interpretability of multi-layered neural networks

International Journal of Advances in Intelligent Informatics

View Publication Info
 
 
Field Value
 
Title Internal and collective interpretation for improving human interpretability of multi-layered neural networks
 
Creator Kamimura, Ryotaro
 
Subject Mutual information; Internal interpretation; Collective interpretation; Inference mechanism; Generalization
 
Description The present paper aims to propose a new type of information-theoretic method to interpret the inference mechanism of neural networks. We interpret the internal inference mechanism for itself without any external methods such as symbolic or fuzzy rules. In addition, we make interpretation processes as stable as possible. This means that we interpret the inference mechanism, considering all internal representations, created by those different conditions and patterns. To make the internal interpretation possible, we try to compress multi-layered neural networks into the simplest ones without hidden layers. Then, the natural information loss in the process of compression is complemented by the introduction of a mutual information augmentation component. The method was applied to two data sets, namely, the glass data set and the pregnancy data set. In both data sets, information augmentation and compression methods could improve generalization performance. In addition, compressed or collective weights from the multi-layered networks tended to produce weights, ironically, similar to the linear correlation coefficients between inputs and targets, while the conventional methods such as the logistic regression analysis failed to do so.
 
Publisher Universitas Ahmad Dahlan
 
Contributor
 
Date 2019-10-29
 
Type info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion

 
Format application/pdf
 
Identifier http://ijain.org/index.php/IJAIN/article/view/420
10.26555/ijain.v5i3.420
 
Source International Journal of Advances in Intelligent Informatics; Vol 5, No 3 (2019): November 2019; 179-192
2548-3161
2442-6571
 
Language eng
 
Relation http://ijain.org/index.php/IJAIN/article/view/420/ijain_v5i3_p179-192
 
Rights https://creativecommons.org/licenses/by-sa/4.0
 

Contact Us

The PKP Index is an initiative of the Public Knowledge Project.

For PKP Publishing Services please use the PKP|PS contact form.

For support with PKP software we encourage users to consult our wiki for documentation and search our support forums.

For any other correspondence feel free to contact us using the PKP contact form.

Find Us

Twitter

Copyright © 2015-2018 Simon Fraser University Library