Evolving Image Classification Architectures With Enhanced Particle Swarm Optimisation
Convolutional Neural Networks (CNNs) have become the de facto technique for image feature extraction in recent years. However, their design and construction remains a complicated task. As more developments are made in progressing the internal components of CNNs, the task of assembling them effective...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2018-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8533601/ |
| _version_ | 1819162222307311616 |
|---|---|
| author | Ben Fielding Li Zhang |
| author_facet | Ben Fielding Li Zhang |
| author_sort | Ben Fielding |
| collection | DOAJ |
| description | Convolutional Neural Networks (CNNs) have become the de facto technique for image feature extraction in recent years. However, their design and construction remains a complicated task. As more developments are made in progressing the internal components of CNNs, the task of assembling them effectively from core components becomes even more arduous. To overcome these barriers, we propose the Swarm Optimized Block Architecture, combined with an enhanced adaptive particle swarm optimization (PSO) algorithm for deep CNN model evolution. The enhanced PSO model employs adaptive acceleration coefficients generated using several cosine annealing mechanisms to overcome stagnation. Specifically, we propose a combined training and structure optimization process for deep CNN model generation, where the proposed PSO model is utilized to explore a bespoke search space defined by a simplified block-based structure. The proposed PSO model not only devises deep networks specifically for image classification, but also builds and pre-trains models for transfer learning tasks. To significantly reduce the hardware and computational cost of the search, the devised CNN model is optimized and trained simultaneously, using a weight sharing mechanism and a final fine-tuning process. Our system compares favorably with related research for optimized deep network generation. It achieves an error rate of 4.78% on the CIFAR-10 image classification task, with 34 hours of combined optimization and training, and an error rate of 25.42% on the CIFAR-100 image data set in 36 hours. All experiments were performed on a single NVIDIA GTX 1080Ti consumer GPU. |
| first_indexed | 2024-12-22T17:24:48Z |
| format | Article |
| id | doaj.art-0335a6c3f1a349d19a98811fee7500a8 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-22T17:24:48Z |
| publishDate | 2018-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-0335a6c3f1a349d19a98811fee7500a82022-12-21T18:18:44ZengIEEEIEEE Access2169-35362018-01-016685606857510.1109/ACCESS.2018.28804168533601Evolving Image Classification Architectures With Enhanced Particle Swarm OptimisationBen Fielding0https://orcid.org/0000-0002-6206-9033Li Zhang1https://orcid.org/0000-0001-6674-692XDepartment of Computer and Information Sciences, Faculty of Engineering and Environment, Northumbria University, Newcastle upon Tyne, U.K.Department of Computer and Information Sciences, Faculty of Engineering and Environment, Northumbria University, Newcastle upon Tyne, U.K.Convolutional Neural Networks (CNNs) have become the de facto technique for image feature extraction in recent years. However, their design and construction remains a complicated task. As more developments are made in progressing the internal components of CNNs, the task of assembling them effectively from core components becomes even more arduous. To overcome these barriers, we propose the Swarm Optimized Block Architecture, combined with an enhanced adaptive particle swarm optimization (PSO) algorithm for deep CNN model evolution. The enhanced PSO model employs adaptive acceleration coefficients generated using several cosine annealing mechanisms to overcome stagnation. Specifically, we propose a combined training and structure optimization process for deep CNN model generation, where the proposed PSO model is utilized to explore a bespoke search space defined by a simplified block-based structure. The proposed PSO model not only devises deep networks specifically for image classification, but also builds and pre-trains models for transfer learning tasks. To significantly reduce the hardware and computational cost of the search, the devised CNN model is optimized and trained simultaneously, using a weight sharing mechanism and a final fine-tuning process. Our system compares favorably with related research for optimized deep network generation. It achieves an error rate of 4.78% on the CIFAR-10 image classification task, with 34 hours of combined optimization and training, and an error rate of 25.42% on the CIFAR-100 image data set in 36 hours. All experiments were performed on a single NVIDIA GTX 1080Ti consumer GPU.https://ieeexplore.ieee.org/document/8533601/Computer visionconvolutional neural networksdeep learningevolutionary computationimage classificationparticle swarm optimization |
| spellingShingle | Ben Fielding Li Zhang Evolving Image Classification Architectures With Enhanced Particle Swarm Optimisation IEEE Access Computer vision convolutional neural networks deep learning evolutionary computation image classification particle swarm optimization |
| title | Evolving Image Classification Architectures With Enhanced Particle Swarm Optimisation |
| title_full | Evolving Image Classification Architectures With Enhanced Particle Swarm Optimisation |
| title_fullStr | Evolving Image Classification Architectures With Enhanced Particle Swarm Optimisation |
| title_full_unstemmed | Evolving Image Classification Architectures With Enhanced Particle Swarm Optimisation |
| title_short | Evolving Image Classification Architectures With Enhanced Particle Swarm Optimisation |
| title_sort | evolving image classification architectures with enhanced particle swarm optimisation |
| topic | Computer vision convolutional neural networks deep learning evolutionary computation image classification particle swarm optimization |
| url | https://ieeexplore.ieee.org/document/8533601/ |
| work_keys_str_mv | AT benfielding evolvingimageclassificationarchitectureswithenhancedparticleswarmoptimisation AT lizhang evolvingimageclassificationarchitectureswithenhancedparticleswarmoptimisation |