TY - GEN
T1 - Optimization of convolutional neural network using microcanonical annealing algorithm
AU - Ayumi, Vina
AU - Rere, L. M.Rasdi
AU - Fanany, Mohamad Ivan
AU - Arymurthy, Aniati Murni
N1 - Funding Information:
This work is supported by Student Paper in Indexed Publication Grant funded by Directorate of Research and Public Services, Universitas Indonesia. Contract No.1854/UN2.R12/HKP.05.0012016.
Publisher Copyright:
© 2016 IEEE.
PY - 2017/3/6
Y1 - 2017/3/6
N2 - Convolutional neural network (CNN) is one of the most prominent architectures and algorithm in Deep Learning. It shows a remarkable improvement in the recognition and classification of objects. This method has also been proven to be very effective in a variety of computer vision and machine learning. As in other deep learning, however, training this approach is interesting yet challenging. Recently, some metaheuristic algorithms have been used to optimize CNN using Genetic Algorithm, Particle Swarm Optimization, Simulated Annealing and Harmony Search. In this paper, another type of metaheuristic algorithms with different strategy has been proposed, i.e. Microcanonical Annealing to optimize Convolutional Neural Network. The performance of the proposed method is tested using the MNIST and CIFAR-10 datasets. Although experiment results of MNIST dataset indicate the increase in computation time (1.02x-1.38x), nevertheless this proposed method can considerably enhance the performance of the original CNN (up to 4.60%). On the CIFAR10 dataset, currently, state of the art is 96.53% using fractional pooling, while this proposed method achieves 99.14%.
AB - Convolutional neural network (CNN) is one of the most prominent architectures and algorithm in Deep Learning. It shows a remarkable improvement in the recognition and classification of objects. This method has also been proven to be very effective in a variety of computer vision and machine learning. As in other deep learning, however, training this approach is interesting yet challenging. Recently, some metaheuristic algorithms have been used to optimize CNN using Genetic Algorithm, Particle Swarm Optimization, Simulated Annealing and Harmony Search. In this paper, another type of metaheuristic algorithms with different strategy has been proposed, i.e. Microcanonical Annealing to optimize Convolutional Neural Network. The performance of the proposed method is tested using the MNIST and CIFAR-10 datasets. Although experiment results of MNIST dataset indicate the increase in computation time (1.02x-1.38x), nevertheless this proposed method can considerably enhance the performance of the original CNN (up to 4.60%). On the CIFAR10 dataset, currently, state of the art is 96.53% using fractional pooling, while this proposed method achieves 99.14%.
KW - CIFAR10
KW - Convolutional Neural Network
KW - MNIST
KW - Metaheuristic
KW - Microcanonical Annealing
UR - http://www.scopus.com/inward/record.url?scp=85017023031&partnerID=8YFLogxK
U2 - 10.1109/ICACSIS.2016.7872787
DO - 10.1109/ICACSIS.2016.7872787
M3 - Conference contribution
AN - SCOPUS:85017023031
T3 - 2016 International Conference on Advanced Computer Science and Information Systems, ICACSIS 2016
SP - 506
EP - 511
BT - 2016 International Conference on Advanced Computer Science and Information Systems, ICACSIS 2016
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 8th International Conference on Advanced Computer Science and Information Systems, ICACSIS 2016
Y2 - 15 October 2016 through 16 October 2016
ER -