Evolutionary Convolutional Neural Network Optimization with Cross-Tasks Transfer Strategy
Convolutional neural networks (CNNs) have shown great success in a variety of real-world applications and the outstanding performance of the state-of-the-art CNNs is primarily driven by the elaborate architecture. Evolutionary convolutional neural network (ECNN) is a promising approach to design the...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-08-01
|
Series: | Electronics |
Subjects: | |
Online Access: | https://www.mdpi.com/2079-9292/10/15/1857 |
_version_ | 1797525670655426560 |
---|---|
author | Zhao Wang Di Lu Huabing Wang Tongfei Liu Peng Li |
author_facet | Zhao Wang Di Lu Huabing Wang Tongfei Liu Peng Li |
author_sort | Zhao Wang |
collection | DOAJ |
description | Convolutional neural networks (CNNs) have shown great success in a variety of real-world applications and the outstanding performance of the state-of-the-art CNNs is primarily driven by the elaborate architecture. Evolutionary convolutional neural network (ECNN) is a promising approach to design the optimal CNN architecture automatically. Nevertheless, most of the existing ECNN methods only focus on improving the performance of the discovered CNN architectures without considering the relevance between different classification tasks. Transfer learning is a human-like learning approach and has been introduced to solve complex problems in the domain of evolutionary algorithms (EAs). In this paper, an effective ECNN optimization method with cross-tasks transfer strategy (CTS) is proposed to facilitate the evolution process. The proposed method is then evaluated on benchmark image classification datasets as a case study. The experimental results show that the proposed method can not only speed up the evolutionary process significantly but also achieve competitive classification accuracy. To be specific, our proposed method can reach the same accuracy at least 40 iterations early and an improvement of accuracy for 0.88% and 3.12% on MNIST-FASHION and CIFAR10 datasets compared with ECNN, respectively. |
first_indexed | 2024-03-10T09:17:17Z |
format | Article |
id | doaj.art-b4e6d4f5873d4d3192b7c66662fa84cc |
institution | Directory Open Access Journal |
issn | 2079-9292 |
language | English |
last_indexed | 2024-03-10T09:17:17Z |
publishDate | 2021-08-01 |
publisher | MDPI AG |
record_format | Article |
series | Electronics |
spelling | doaj.art-b4e6d4f5873d4d3192b7c66662fa84cc2023-11-22T05:32:00ZengMDPI AGElectronics2079-92922021-08-011015185710.3390/electronics10151857Evolutionary Convolutional Neural Network Optimization with Cross-Tasks Transfer StrategyZhao Wang0Di Lu1Huabing Wang2Tongfei Liu3Peng Li4State Key Laboratory of Complex Electromagnetic Environment Effects on Electronics and Information System (CEMEE), Luoyang 471003, ChinaKey Laboratory of Electronic Information Countermeasure and Simulation Technology of Ministry of Education, Xidian University, No. 2 South TaiBai Road, Xi’an 710071, ChinaState Key Laboratory of Complex Electromagnetic Environment Effects on Electronics and Information System (CEMEE), Luoyang 471003, ChinaKey Laboratory of Electronic Information Countermeasure and Simulation Technology of Ministry of Education, Xidian University, No. 2 South TaiBai Road, Xi’an 710071, ChinaKey Laboratory of Electronic Information Countermeasure and Simulation Technology of Ministry of Education, Xidian University, No. 2 South TaiBai Road, Xi’an 710071, ChinaConvolutional neural networks (CNNs) have shown great success in a variety of real-world applications and the outstanding performance of the state-of-the-art CNNs is primarily driven by the elaborate architecture. Evolutionary convolutional neural network (ECNN) is a promising approach to design the optimal CNN architecture automatically. Nevertheless, most of the existing ECNN methods only focus on improving the performance of the discovered CNN architectures without considering the relevance between different classification tasks. Transfer learning is a human-like learning approach and has been introduced to solve complex problems in the domain of evolutionary algorithms (EAs). In this paper, an effective ECNN optimization method with cross-tasks transfer strategy (CTS) is proposed to facilitate the evolution process. The proposed method is then evaluated on benchmark image classification datasets as a case study. The experimental results show that the proposed method can not only speed up the evolutionary process significantly but also achieve competitive classification accuracy. To be specific, our proposed method can reach the same accuracy at least 40 iterations early and an improvement of accuracy for 0.88% and 3.12% on MNIST-FASHION and CIFAR10 datasets compared with ECNN, respectively.https://www.mdpi.com/2079-9292/10/15/1857evolutionary algorithmconvolutional neural networktransfer learningimage classification |
spellingShingle | Zhao Wang Di Lu Huabing Wang Tongfei Liu Peng Li Evolutionary Convolutional Neural Network Optimization with Cross-Tasks Transfer Strategy Electronics evolutionary algorithm convolutional neural network transfer learning image classification |
title | Evolutionary Convolutional Neural Network Optimization with Cross-Tasks Transfer Strategy |
title_full | Evolutionary Convolutional Neural Network Optimization with Cross-Tasks Transfer Strategy |
title_fullStr | Evolutionary Convolutional Neural Network Optimization with Cross-Tasks Transfer Strategy |
title_full_unstemmed | Evolutionary Convolutional Neural Network Optimization with Cross-Tasks Transfer Strategy |
title_short | Evolutionary Convolutional Neural Network Optimization with Cross-Tasks Transfer Strategy |
title_sort | evolutionary convolutional neural network optimization with cross tasks transfer strategy |
topic | evolutionary algorithm convolutional neural network transfer learning image classification |
url | https://www.mdpi.com/2079-9292/10/15/1857 |
work_keys_str_mv | AT zhaowang evolutionaryconvolutionalneuralnetworkoptimizationwithcrosstaskstransferstrategy AT dilu evolutionaryconvolutionalneuralnetworkoptimizationwithcrosstaskstransferstrategy AT huabingwang evolutionaryconvolutionalneuralnetworkoptimizationwithcrosstaskstransferstrategy AT tongfeiliu evolutionaryconvolutionalneuralnetworkoptimizationwithcrosstaskstransferstrategy AT pengli evolutionaryconvolutionalneuralnetworkoptimizationwithcrosstaskstransferstrategy |