TY - GEN

T1 - Modification of architecture learning convolutional neural network for graph

AU - Rukmanda, T. D.

AU - Sugeng, K. A.

AU - Murfi, H.

N1 - Funding Information:
Part of this research has been funded by PITTA Universitas Indonesia Research Grant 2017 with contract number 662/UN2.R3.1/HKP.05.00/2017.
Publisher Copyright:
© 2018 Author(s).

PY - 2018/10/22

Y1 - 2018/10/22

N2 - 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.

AB - 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.

KW - convolutional neural networks

KW - deep learning

KW - graph normalization

UR - http://www.scopus.com/inward/record.url?scp=85056108545&partnerID=8YFLogxK

U2 - 10.1063/1.5064197

DO - 10.1063/1.5064197

M3 - Conference contribution

AN - SCOPUS:85056108545

T3 - AIP Conference Proceedings

BT - Proceedings of the 3rd International Symposium on Current Progress in Mathematics and Sciences 2017, ISCPMS 2017

A2 - Yuniati, Ratna

A2 - Mart, Terry

A2 - Anggraningrum, Ivandini T.

A2 - Triyono, Djoko

A2 - Sugeng, Kiki A.

PB - American Institute of Physics Inc.

T2 - 3rd International Symposium on Current Progress in Mathematics and Sciences 2017, ISCPMS 2017

Y2 - 26 July 2017 through 27 July 2017

ER -