|
- 2018
混合PSO优化卷积神经网络结构和参数
|
Abstract:
为了使卷积神经网络在非经验指导下自动寻得最优连接,并提高其参数优化效率,提出用粒子群优化卷积网络参数,并用离散粒子群优化卷积网络特征图之间连接结构的新方法。先使用粒子群优化所有权值,再采用离散粒子群优化降采样层和卷积层之间特征图连接结构。将该方法用于MNIST数据集和CIFAR-10数据集,实验结果表明,相比其他连接结构的卷积神经网络和其他识别方法,该方法可以有效实现网络结构及参数的优化,加速网络收敛并提高识别准确比。
[1] | CLEVERT D A, UNTERTHINER T, HOCHREITER S. Fast and accurate deep network learning by exponential linear units (ELUs)[C]//International Conference on Learning Representations. San Juan:Computer Science, 2016:arXiv:1511.07289. |
[2] | ARORA C, SABETZADEH M, BRIAND L, et al. Automated checking of conformance to requirements templates using natural language processing[J]. IEEE Transactions on Software Engineering, 2015, 41(10):944-968. |
[3] | ZHANG Zhong, WANG Chun-heng, XIAO Bai-hua, et al. Cross-view action recognition using contextual maximum margin clustering[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2014, 24(10):1663-1668. |
[4] | LECUN Y, BOSER B, DENKER J S, et al. Backpropagation applied to handwritten zip code recognition[J]. Neural Computation, 1989, 1(4):541-551. |
[5] | LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11):2278-2324. |
[6] | YUAN Q, YIN G. Analyzing convergence and rates of convergence of particle swarm optimization algorithms using stochastic approximation methods[J]. IEEE Transactions on Automatic Control, 2015, 60(7):1760-1773. |
[7] | OULLRTTE R, BROWNE M, HIRASAWA K. Genetic algorithm optimization of a convolutional neuralnetwork for autonomous crack detection[C]//Congress on Evolutionary Computation. Portland:IEEE, 2004, 1:516-521. |
[8] | XIONG Shi-fu, WU Guo, LIU Di-yuan. The Vietnamese speech recognition based on rectified linear units deep neural network and spoken term detection system combination[C]//9th International Symposium on Chinese Spoken Language Processing (ISCSLP). Singapore:IEEE, 2014:183-186. |
[9] | LI Y H, ZHAN Z H, LIN S J, et al. Competitive and cooperative particle swarm optimization with information sharing mechanism for global optimization problems[J]. Information Sciences, 2015, 293:370-382. |
[10] | LECUN Y, CORTES C, BURGES C J C. The MNIST database of handwritten digits[EB/OL].[2016-07-12]. http://yann.lecun.com/exdb/mnist/. |
[11] | LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 2015, 215:436-444. |
[12] | 戴晓爱, 郭守恒, 任淯, 等. 基于堆栈式稀疏自编码器的高光谱影像分类[J]. 电子科技大学学报, 2016, 45(3):382-386. DAI Xiao-ai, GUO Shou-heng, REN Yu, et al. Hyperspectral remote sensing image classification using the stacked sparse autoencoder[J]. Journal of University of Electronic Science and Technology of China, 2016, 45(3):382-386. |
[13] | FEDOROVICI L O, PRECUP R E, DRAGAN F, et al. Evolutionary optimization-based training of convolutional neural networks for OCR applications[C]//201317th International Conference on System Theory, Control and Computing (ICSTCC). Sinaia:IEEE, 2013:207-212. |
[14] | LIU Dong, JIANG Qi-long, CHEN J X. Binary inheritance learning particle swarm optimisation and its application in thinned antenna array synthesis with the minimum sidelobe level[J]. IET Microwaves, Antennas & Propagation, 2015, 9(13):1386-1391. |
[15] | 余萍, 赵继生, 张洁. 基于非线性修正函数的卷积神经网络图像识别研究[J]. 科学技术与工程, 2015, 15(34):221-225. YU Ping, ZHAO Ji-sheng, ZHANG Jie. Image recognition of convolutional neural networks based on rectified nonlinear units function[J]. Science Technology and Engineering, 2015, 15(34):221-225. |