Modification of architecture learning convolutional neural network for graph

T. D. Rukmanda, Kiki Ariyanti, Hendri Murfi

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

PATCHY-SAN is a framework for learning Convolutional Neural Network (CNNs) for an arbitrary graph. In this paper, we propose a modified architecture of Convolutional Neural Network in PATCHY-SAN. We use some representation matrices of a graph such as Bi, Li, Ni, with B, L, N, are a betweenness matrix, a Laplacian matrix and a normalize Laplacian matrix with i = 1, 2, 3, 4, 5. We do some experiment of a model with three convolutional layers and two convolutional layers. In order to reduce internal covariate shift, we use a batch normalization as a regularizer. In conclusion, by adding more convolution layers, and using batch normalization can increase and reduce accuracy. The accuracy is more dependent on the architecture of CNNs.

Original languageEnglish
Title of host publicationProceedings of the 3rd International Symposium on Current Progress in Mathematics and Sciences 2017, ISCPMS 2017
EditorsRatna Yuniati, Terry Mart, Ivandini T. Anggraningrum, Djoko Triyono, Kiki A. Sugeng
PublisherAmerican Institute of Physics Inc.
ISBN (Electronic)9780735417410
DOIs
Publication statusPublished - 22 Oct 2018
Event3rd International Symposium on Current Progress in Mathematics and Sciences 2017, ISCPMS 2017 - Bali, Indonesia
Duration: 26 Jul 201727 Jul 2017

Publication series

NameAIP Conference Proceedings
Volume2023
ISSN (Print)0094-243X
ISSN (Electronic)1551-7616

Conference

Conference3rd International Symposium on Current Progress in Mathematics and Sciences 2017, ISCPMS 2017
CountryIndonesia
CityBali
Period26/07/1727/07/17

Keywords

  • convolutional neural networks
  • deep learning
  • graph normalization

Fingerprint Dive into the research topics of 'Modification of architecture learning convolutional neural network for graph'. Together they form a unique fingerprint.

Cite this