TY - GEN
T1 - Least square adversarial autoencoder
AU - Sinaga, Marshal Anjona
AU - Stefanus, Lim Yohanes
N1 - Funding Information:
We gratefully acknowledge the support of the Tokopedia-UI AI Center of Excellence, Faculty of Computer Science, University of Indonesia, for allowing us to use its NVIDIA DGX-1 for running our experiments.
Publisher Copyright:
© 2020 IEEE.
Copyright:
Copyright 2021 Elsevier B.V., All rights reserved.
PY - 2020/10/17
Y1 - 2020/10/17
N2 - This research introduces least square adversarial autoencoder (LSAA)-an autoencoder that is able to reconstruct data and also generate data that has characteristics similar to data distribution from the prior distribution. LSAA uses least square generative adversarial network loss function on its discriminator. LSAA minimizes Pearson ?2 divergence between the latent variable distribution and the prior distribution. In this research, a Python program is developed to model LSAA by utilizing MNIST data set and FashionMNIST data set. The program is implemented using PyTorch. All of the programming activities are carried out in the cloud environment provided by the Tokopedia-Universitas Indonesia AI Center, using DGX-I (GPU Tesla V100) as its computing resource. The experimental results show that the mean squared error of LSAA for MNIST data set and FashionMNIST data set are 0.0080 and 0.0099, respectively. Furthermore, the Fréchet Inception Distance score of LSAA for MNIST data set and FashionMNIST data set are 11.1280 and 27.5737, respectively. These results indicate that the least square adversarial autoencoder is able to reconstruct the image properly and also able to generate images similar to the training samples.
AB - This research introduces least square adversarial autoencoder (LSAA)-an autoencoder that is able to reconstruct data and also generate data that has characteristics similar to data distribution from the prior distribution. LSAA uses least square generative adversarial network loss function on its discriminator. LSAA minimizes Pearson ?2 divergence between the latent variable distribution and the prior distribution. In this research, a Python program is developed to model LSAA by utilizing MNIST data set and FashionMNIST data set. The program is implemented using PyTorch. All of the programming activities are carried out in the cloud environment provided by the Tokopedia-Universitas Indonesia AI Center, using DGX-I (GPU Tesla V100) as its computing resource. The experimental results show that the mean squared error of LSAA for MNIST data set and FashionMNIST data set are 0.0080 and 0.0099, respectively. Furthermore, the Fréchet Inception Distance score of LSAA for MNIST data set and FashionMNIST data set are 11.1280 and 27.5737, respectively. These results indicate that the least square adversarial autoencoder is able to reconstruct the image properly and also able to generate images similar to the training samples.
KW - Artificial neural network
KW - Autoencoder
KW - Generative model
KW - Least square generative adversarial network
KW - Regularization
UR - http://www.scopus.com/inward/record.url?scp=85099742409&partnerID=8YFLogxK
U2 - 10.1109/ICACSIS51025.2020.9263181
DO - 10.1109/ICACSIS51025.2020.9263181
M3 - Conference contribution
AN - SCOPUS:85099742409
T3 - 2020 International Conference on Advanced Computer Science and Information Systems, ICACSIS 2020
SP - 33
EP - 40
BT - 2020 International Conference on Advanced Computer Science and Information Systems, ICACSIS 2020
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 12th International Conference on Advanced Computer Science and Information Systems, ICACSIS 2020
Y2 - 17 October 2020 through 18 October 2020
ER -