Convolutional Neural Network Architecture Design by the Tree Growth Algorithm Framework

Ivana Strumberger, Eva Tuba, Nebojsa Bacanin, Raka Jovanovic, Milan Tuba

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

This paper presents tree growth algorithm framework for designing convolutional neural network architecture. Convolutional neural networks are a special class of deep neural networks that typically consist of several convolution, pooling and fully connected layers. Convolutional neural networks have proved to be a robust method for tackling various image classification tasks. One of the most important challenges from this domain is to find the network architecture that has the best performance for the specific application. The performance of the network depends on the set of hyper-parameter values such as the number of convolutional and dense layers, the number of kernels per layer and kernel size. Optimization of hyperparameters was performed by novel tree growth algorithm that belongs to the group of swarm intelligence metaheuristics. The robustness, performance and solutions quality of the proposed framework was validated against the well-known MNIST dataset. Conducted comparative analysis demonstrated that the proposed frameworks obtains promising results in this domain.

Original languageEnglish
Title of host publication2019 International Joint Conference on Neural Networks, IJCNN 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728119854
DOIs
Publication statusPublished - Jul 2019
Event2019 International Joint Conference on Neural Networks, IJCNN 2019 - Budapest, Hungary
Duration: 14 Jul 201919 Jul 2019

Publication series

NameProceedings of the International Joint Conference on Neural Networks
Volume2019-July

Conference

Conference2019 International Joint Conference on Neural Networks, IJCNN 2019
CountryHungary
CityBudapest
Period14/7/1919/7/19

Fingerprint

Network architecture
Trees (mathematics)
Neural networks
Image classification
Convolution

Keywords

  • covolutional neural networks
  • hyper-parameters
  • optimization
  • swarm intelligence
  • tree growth algorithm

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence

Cite this

Strumberger, I., Tuba, E., Bacanin, N., Jovanovic, R., & Tuba, M. (2019). Convolutional Neural Network Architecture Design by the Tree Growth Algorithm Framework. In 2019 International Joint Conference on Neural Networks, IJCNN 2019 [8851755] (Proceedings of the International Joint Conference on Neural Networks; Vol. 2019-July). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IJCNN.2019.8851755

Convolutional Neural Network Architecture Design by the Tree Growth Algorithm Framework. / Strumberger, Ivana; Tuba, Eva; Bacanin, Nebojsa; Jovanovic, Raka; Tuba, Milan.

2019 International Joint Conference on Neural Networks, IJCNN 2019. Institute of Electrical and Electronics Engineers Inc., 2019. 8851755 (Proceedings of the International Joint Conference on Neural Networks; Vol. 2019-July).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Strumberger, I, Tuba, E, Bacanin, N, Jovanovic, R & Tuba, M 2019, Convolutional Neural Network Architecture Design by the Tree Growth Algorithm Framework. in 2019 International Joint Conference on Neural Networks, IJCNN 2019., 8851755, Proceedings of the International Joint Conference on Neural Networks, vol. 2019-July, Institute of Electrical and Electronics Engineers Inc., 2019 International Joint Conference on Neural Networks, IJCNN 2019, Budapest, Hungary, 14/7/19. https://doi.org/10.1109/IJCNN.2019.8851755
Strumberger I, Tuba E, Bacanin N, Jovanovic R, Tuba M. Convolutional Neural Network Architecture Design by the Tree Growth Algorithm Framework. In 2019 International Joint Conference on Neural Networks, IJCNN 2019. Institute of Electrical and Electronics Engineers Inc. 2019. 8851755. (Proceedings of the International Joint Conference on Neural Networks). https://doi.org/10.1109/IJCNN.2019.8851755
Strumberger, Ivana ; Tuba, Eva ; Bacanin, Nebojsa ; Jovanovic, Raka ; Tuba, Milan. / Convolutional Neural Network Architecture Design by the Tree Growth Algorithm Framework. 2019 International Joint Conference on Neural Networks, IJCNN 2019. Institute of Electrical and Electronics Engineers Inc., 2019. (Proceedings of the International Joint Conference on Neural Networks).
@inproceedings{698818f640b2401c9b9e1281d2ec8251,
title = "Convolutional Neural Network Architecture Design by the Tree Growth Algorithm Framework",
abstract = "This paper presents tree growth algorithm framework for designing convolutional neural network architecture. Convolutional neural networks are a special class of deep neural networks that typically consist of several convolution, pooling and fully connected layers. Convolutional neural networks have proved to be a robust method for tackling various image classification tasks. One of the most important challenges from this domain is to find the network architecture that has the best performance for the specific application. The performance of the network depends on the set of hyper-parameter values such as the number of convolutional and dense layers, the number of kernels per layer and kernel size. Optimization of hyperparameters was performed by novel tree growth algorithm that belongs to the group of swarm intelligence metaheuristics. The robustness, performance and solutions quality of the proposed framework was validated against the well-known MNIST dataset. Conducted comparative analysis demonstrated that the proposed frameworks obtains promising results in this domain.",
keywords = "covolutional neural networks, hyper-parameters, optimization, swarm intelligence, tree growth algorithm",
author = "Ivana Strumberger and Eva Tuba and Nebojsa Bacanin and Raka Jovanovic and Milan Tuba",
year = "2019",
month = "7",
doi = "10.1109/IJCNN.2019.8851755",
language = "English",
series = "Proceedings of the International Joint Conference on Neural Networks",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "2019 International Joint Conference on Neural Networks, IJCNN 2019",

}

TY - GEN

T1 - Convolutional Neural Network Architecture Design by the Tree Growth Algorithm Framework

AU - Strumberger, Ivana

AU - Tuba, Eva

AU - Bacanin, Nebojsa

AU - Jovanovic, Raka

AU - Tuba, Milan

PY - 2019/7

Y1 - 2019/7

N2 - This paper presents tree growth algorithm framework for designing convolutional neural network architecture. Convolutional neural networks are a special class of deep neural networks that typically consist of several convolution, pooling and fully connected layers. Convolutional neural networks have proved to be a robust method for tackling various image classification tasks. One of the most important challenges from this domain is to find the network architecture that has the best performance for the specific application. The performance of the network depends on the set of hyper-parameter values such as the number of convolutional and dense layers, the number of kernels per layer and kernel size. Optimization of hyperparameters was performed by novel tree growth algorithm that belongs to the group of swarm intelligence metaheuristics. The robustness, performance and solutions quality of the proposed framework was validated against the well-known MNIST dataset. Conducted comparative analysis demonstrated that the proposed frameworks obtains promising results in this domain.

AB - This paper presents tree growth algorithm framework for designing convolutional neural network architecture. Convolutional neural networks are a special class of deep neural networks that typically consist of several convolution, pooling and fully connected layers. Convolutional neural networks have proved to be a robust method for tackling various image classification tasks. One of the most important challenges from this domain is to find the network architecture that has the best performance for the specific application. The performance of the network depends on the set of hyper-parameter values such as the number of convolutional and dense layers, the number of kernels per layer and kernel size. Optimization of hyperparameters was performed by novel tree growth algorithm that belongs to the group of swarm intelligence metaheuristics. The robustness, performance and solutions quality of the proposed framework was validated against the well-known MNIST dataset. Conducted comparative analysis demonstrated that the proposed frameworks obtains promising results in this domain.

KW - covolutional neural networks

KW - hyper-parameters

KW - optimization

KW - swarm intelligence

KW - tree growth algorithm

UR - http://www.scopus.com/inward/record.url?scp=85073199094&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073199094&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2019.8851755

DO - 10.1109/IJCNN.2019.8851755

M3 - Conference contribution

AN - SCOPUS:85073199094

T3 - Proceedings of the International Joint Conference on Neural Networks

BT - 2019 International Joint Conference on Neural Networks, IJCNN 2019

PB - Institute of Electrical and Electronics Engineers Inc.

ER -