计算机科学 ›› 2018, Vol. 45 ›› Issue (1): 280-284.doi: 10.11896/j.issn.1002-137X.2018.01.049
任俊,胡晓峰,李宁
REN Jun, HU Xiao-feng and LI Ning
摘要: 为了解决大数据时代下小样本数据预测精度不高的问题,提出一种基于堆栈降噪自编码(SDA)与支持向量回归机(SVR)的混合模型。该方法采用源域大样本数据对堆栈降噪自编码和支持向量回归机混合模型进行迁移预训练,再利用目标域小样本数据微调混合模型。堆栈降噪自编码器具有良好的通用深层特征自主抽取能力,能够发掘源领域与目标领域相似任务间的共有特征知识,该知识能够辅助支持向量回归机在高维噪声小样本数据集上的预测。在多种数据集上的实验结果证明了该方法的有效性。
[1] TAO J W,CHUNG F L,WANG S T.A kernel learning framework for domain adaptation learning[J].Science China Information Sciences,2012,5(9):1983-2007. [2] PAN J L,YANG Q.A survey on transfer learning[J].IEEETrans.on knowledge and Data Engineering,2010,2(10):1345-1359. [3] PANS J L,KWOK J T,YANG Q.Transferring learning viadimensionality reduction[C]∥Proc.of the 23st National Con-ference on Artificial Intelligence.Menlo Park:AAAI Press,2008:677-682. [4] WEI F M,ZHANG J P,CHU Y,et al.FSFP:Transfer learning from long texts to the short[J].Applied Mathematics & Information Sciences,2014,8(4):2033-2044. [5] OQUAB M M,BOTTOU L,LAPTEV I,et al.Learning andTransferring Mid-level Image Representations Using Convolutional Neural Networks [C]∥2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).USA:IEEE Computer Society,2014:1717-l724. [6] AREL I,ROSE D C,KARNOWSKI T P.Deep Machine Lear-ning-A new Frontier in Artificial intelligence research[J].IEEE Computational Intelligence Magazine,2010,5(4):13-18. [7] LECUN Y,BENGIO Y,HINTON G.Deep learning[J].Nature,2015,521(5):436-444. [8] YU K,JIA L,CHEN Y Q,et al.Deep Learning.Yesterday,Today,and Tomorrow[J].Journal of Computer Research and Dve-lopment,2013,0(9):1799-1804.(in Chinese) 余凯,贾磊,陈雨强,等.深度学习的昨天、今天和明天[J].计算机研究与发展,2013,0(9):1799-1804. [9] BENGIO Y .Learning deep architectures for AI[J].Foundations and Trends in Machine Learning,2009,2(1):1-127. [10] HINTON G,SALAKHUTDINOV R R.Reducing the dimen-sionality of data with neural networks[J].Science,2006,313(5786):504-507. [11] ZHANG L,LIU Z,ZHANG J Q,et al.Reduction method of weapon system-of-systems assessment index system based on autoencoder[J].Journal of Central South University(Science and Technology),2013,4(10):4130-4137.(in Chinese) 张乐,刘忠,张建强,等.基于自编码神经网络的装备体系评估指标约简方法[J].中南大学学报(自然科学版),2013,4(10):4130-4137. [12] HU Z,FU K,ZHANG C S.Audio Classical Composer Identification by Deep Neural Network[J].Journal of Computer Research and Development,2014,1(9):1945-1954.(in Chinese) 胡振,傅昆,张长水.基于深度学习的作曲家分类问题[J].计算机研究与发展,2014,1(9):1945-1954. [13] SONG C F,LIU F,HUANG Y Z,et al.Auto-encoder based Data clustering[C]∥Proceeding,Part I,of the 18th Iberoamerican Congress on Progress in Pattern Recognition,Image Analysis,Computer Vision,and Applications.Springer-verlag:New York,2013:117-124. [14] VINCENT P,LAROCHELLE H,BENGIO Y ,et al.Extractingand composing robust features with denoising autoeneoders[C]∥ Proceedings of the 25th International Conference on Machine Learning.New York,NY,USA:ACM,2008:1096-1103. [15] BENGIO Y,LAMBLIN P,LAROCHELLE H,et al.Greedylayerwise training of deep networks[J].Advances in Neural Information Processing Systems,2007,19:153-160. [16] GAN M T,HANMANDLU M,TAN A H.From a Gaussianmixture model to additive fuzzy systems[J].IEEE Transactions on Fuzzy Systems,2005,3(3):303-316. |
No related articles found! |
|