Deep transfer adaptation network based on improved maximum mean discrepancy algorithm

doi:10.11772/j.issn.1001-9081.2020020263

Journal of Computer Applications ›› 2020, Vol. 40 ›› Issue (11): 3107-3112.DOI: 10.11772/j.issn.1001-9081.2020020263

• Artificial intelligence • Previous Articles Next Articles

Deep transfer adaptation network based on improved maximum mean discrepancy algorithm

ZHENG Zongsheng¹, HU Chenyu¹, JIANG Xiaoyi²

1. College of Information and Science, Shanghai Ocean University, Shanghai 201306, China;
2. National Marine Information Center, Tianjin 300171, China

Received:2020-03-14 Revised:2020-06-01 Online:2020-11-10 Published:2020-07-07
Supported by:
This work is partially supported by the National Natural Science Foundation of China (41671431), the Shanghai Municipal Science and Technology Commission Local Colleges Capacity Building Project (17050501900), the Open Fund of Key Laboratory of Digital Ocean Science and Technology of State Oceanic Administration.

基于改进的最大均值差异算法的深度迁移适配网络

郑宗生¹, 胡晨雨¹, 姜晓轶²

1. 上海海洋大学信息学院, 上海 201306;
2. 国家海洋信息中心, 天津 300171

通讯作者: 胡晨雨(1995-),女,江苏盐城人,硕士研究生,主要研究方向:迁移学习、图像处理;1105814265@qq.com
作者简介:郑宗生(1979-),男,河北唐山人,副教授,博士,主要研究方向:深度学习、迁移学习、遥感图像处理;姜晓轶(1973-),男,黑龙江齐齐哈尔人,博士,主要研究方向:时空数据库、地理信息系统
基金资助:
国家自然科学基金资助项目（41671431）；上海市科委地方院校能力建设项目（17050501900）；国家海洋局数字海洋科学技术重点实验室开放基金资助项目。

Abstract

Abstract: In the study of model parameter based transfer learning, both the sample distribution discrepancy between two domains and the co-adaptation between convolutional layers of the source model impact performance of model. In response to these problems, a Multi-Convolution Adaptation (MCA) deep transfer framework was proposed and applied to the grade classification of typhoons in satellite cloud images, and a CE-MMD loss function was defined by adding the improved L-MMD (Maximum Mean Discrepancy) algorithm as a regular term to the cross-entropy function and applying the linear unbiased estimation to the distribution of the samples in Reproducing Kernel Hilbert Space (RKHS). In the back propagation process, the residual error and the distribution discrepancy between the samples in two domains were used as common indexes to update the network parameters, making model converge faster and have higher accuracy. Comparison experimental results of L-MMD and two measurement algorithms-Bregman difference and KL (Kullback-Leibler) divergence on the self-built typhoon dataset show that the precision of the proposed algorithm is improved by 11.76 percentage points and 8.05 percentage points respectively compared to those of the other two algorithms. It verifies that L-MMD is superior to other measurement algorithms and the MCA deep transfer framework is feasible.

Key words: transfer learning, Deep Convolutional Neural Network (DCNN), Maximum Mean Discrepancy (MMD), domain adaptation, typhoon grade

摘要： 在基于模型参数的迁移学习研究中，两域样本的分布差异、源模型卷积层之间的互适应性都是影响模型迁移性能的重要因素。针对上述问题，提出一种多层卷积适配（MCA）深度迁移框架并将其应用于台风卫星云图的等级分类，在交叉熵函数的基础上添加L-最大均值差异（MMD）算法作为正则项，并对样本在再生核希尔伯特空间（RKHS）中的分布进行线性的无偏估计，最终定义了CE-MMD损失函数。在反向传播过程中，将残差和两域样本分布的差异共同作为网络参数更新的指标，使得迁移模型收敛速度更快、精度更高。在自建的台风数据集上对L-MMD算法和Bregman差异、KL散度两种度量算法进行对比实验，结果表明所提算法的精度分别高出11.76个百分点和8.05个百分点。实验结果表明，L-MMD算法优于其他度量算法，而且MCA深度迁移框架具有可行性。

关键词: 迁移学习, 深度卷积神经网络, 最大均值差异, 领域适配, 台风等级

CLC Number:

TP181

ZHENG Zongsheng, HU Chenyu, JIANG Xiaoyi. Deep transfer adaptation network based on improved maximum mean discrepancy algorithm[J]. Journal of Computer Applications, 2020, 40(11): 3107-3112.

郑宗生, 胡晨雨, 姜晓轶. 基于改进的最大均值差异算法的深度迁移适配网络[J]. 计算机应用, 2020, 40(11): 3107-3112.

References

[1] 庄福振, 罗平, 何清, 等. 迁移学习研究进展[J]. 软件学报, 2015,26(1):26-39. (ZHUANG F Z,LUO P,HE Q,et al. Survey on transfer learning research[J]. Journal of Software,2015, 26(1):26-39.)
[2] PAN S J, NI X, SUN J T, et al. Cross-domain sentiment classification via spectral feature alignment[C]//Proceedings of the 19th International Conference on World Wide Web. New York:ACM,2010:751-760.
[3] QI G J,AGGARWAL C,HUANG T. Towards semantic knowledge propagation from text corpus to Web images[C]//Proceedings of the 20th International Conference on World Wide Web. New York:ACM,2011:297-306.
[4] LONG M,WANG J,DING G,et al. Transfer joint matching for unsupervised domain adaptation[C]//Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2014:1410-1417.
[5] XIAO M,GUO Y. Feature space independent semi-supervised domain adaptation via kernel matching[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,37(1):54-66.
[6] WANG X,SCHNEIDER J. Flexible transfer learning under support and model shift[C]//Proceedings of the 27th International Conference on Neural Information Processing Systems. Cambridge:MIT Press,2014:1898-1906.
[7] KHAN M N A,HEISTERKAMP D R. Adapting instance weights for unsupervised domain adaptation using quadratic mutual information and subspace learning[C]//Proceedings of the 23rd International Conference on Pattern Recognition. Piscataway:IEEE,2016:1560-1565.
[8] SIMONYAN K,ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[EB/OL].[2020-02-22]. https://arxiv.org/pdf/1409.1556.pdf.
[9] HE K,ZHANG X,REN S,et al. Deep residual learning for image recognition[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2016:770-778.
[10] SZEGEDY C,VANHOUCKE V,IOFFE S,et al. Rethinking the inception architecture forcomputer vision[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2016:2818-2826.
[11] BENGIO Y, COURVILLE A, VINCENT P. Representation learning:a review and new perspectives[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(8):1798-1828.
[12] YOSINSKI J,CLUNE J,BENGIO Y,et al. How transferable are features in deep neural networks?[C]//Proceedings of the 27th International Conference on Neural Information Processing Systems. Cambridge:MIT Press,2014:3320-3328.
[13] LONG M,CAO Y,WANG J,et al. Learning transferable features with deep adaptation networks[C]//Proceedings of the 32nd International Conference on Machine Learning. New York:JMLR.org,2015:97-105.
[14] PAN S J,TSANG I W,KWOK J T,et al. Domain adaptation via transfercomponent analysis[J]. IEEE Transactions on Neural Networks,2011,22(2):199-210.
[15] GRETTON A,SRIPERUMBUDUR B,SEJDINOVIC D,et al. Optimal kernel choice for large-scale two-sample tests[C]//Proceedings of the 25th International Conference on Neural Information Processing Systems. Red Hook, NY:Curran Associates Inc.,2012:1205-1213.
[16] PAN S J,YANG Q. A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering,2010,22(10):1345-1359.
[17] LONG M,WANG J,DING G,et. al. Transfer feature learning with joint distribution adaptation[C]//Proceedings of the 2013 IEEE International Conference on Computer Vision. Piscataway:IEEE,2013:2200-2207.
[18] 郑宗生, 胡晨雨, 黄冬梅, 等. 基于迁移学习及气象卫星云图的台风等级分类研究[J]. 遥感技术与应用,2020,35(1):202-210. (ZHENG Z S,HU C Y,HUANG D M,et al. Research on transfer learning methods for classification of typhoon cloud image[J]. Remote Sensing Technology and Application,2020,35(1):202-210.)
[19] 皋军, 黄丽莉. 最大局部加权均值差异嵌入[J]. 电子学报, 2013,41(8):1462-1468. (GAO J,HUANG L L. Maximum local weighted mean discrepancy embedding[J]. Acta Electronica Sinica,2013,41(8):1462-1468.)
[20] TZENG E, HOFFMAN J, ZHANG N, et al. Deep domain confusion:maximizing for domain invariance[EB/OL].[2020-03-01]. https://arxiv.org/pdf/1412.3474.pdf.
[21] CAO X,WIPF D,WEN F,et al. A practical transfer learning algorithm for face verification[C]//Proceedings of the 2013 IEEE International Conference on Computer Vision. Piscataway:IEEE, 2013:3208-3215.

Deep transfer adaptation network based on improved maximum mean discrepancy algorithm

基于改进的最大均值差异算法的深度迁移适配网络

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics

[1]	CHEN Zhengtao, HUANG Can, YANG Bo, ZHAO Li, LIAO Yong. Yak face recognition algorithm of parallel convolutional neural network based on transfer learning [J]. Journal of Computer Applications, 2021, 41(5): 1332-1336.
[2]	WANG Jinkai, JIA Xu. Vein recognition algorithm based on Siamese nonnegative matrix factorization with transferability [J]. Journal of Computer Applications, 2021, 41(3): 898-903.
[3]	HUANG Xueyu, XU Haote, TAO Jianwen. Multi-source adaptation classification framework with feature selection [J]. Journal of Computer Applications, 2020, 40(9): 2499-2506.
[4]	ZHANG Chenxi, TANG Shu, TANG Ke. Strategies of parameter fault detection for rocket engines based on transfer learning [J]. Journal of Computer Applications, 2020, 40(9): 2774-2780.
[5]	BAI Ruyi, GUO Xiaoying, JIA Chunhua. Sentiment prediction of small sample abstract painting image based on feature fusion [J]. Journal of Computer Applications, 2020, 40(8): 2207-2213.
[6]	YUAN Yuan, WU Wen, WAN Yi. Single image shadow detection method based on entropy driven domain adaptive learning [J]. Journal of Computer Applications, 2020, 40(7): 2131-2136.
[7]	MA Jinlin, WEI Meng, MA Ziping. Pulmonary nodule segmentation method based on deep transfer learning [J]. Journal of Computer Applications, 2020, 40(7): 2117-2125.
[8]	XIAO He, LIU Zhiqin, WANG Qingfeng, HUANG Jun, ZHOU Ying, LIU Qiyu, XU Weiyun. Mass and calcification classification method in mammogram based on multi-view transfer learning [J]. Journal of Computer Applications, 2020, 40(5): 1460-1464.
[9]	CHEN Jie, ZHANG Ting, DU Yi. Reconstruction of porous media using adaptive deep transfer learning [J]. Journal of Computer Applications, 2020, 40(4): 1231-1236.
[10]	CHEN Guihui, YI Xin, LI Zhongbing, QIAN Jiren, CHEN Wu. Third-party construction target detection in aerial images of pipeline inspection based on improved YOLOv2 and transfer learning [J]. Journal of Computer Applications, 2020, 40(4): 1062-1068.
[11]	MAO Wentao, YANG Chao, LIU Yamin, TIAN Siyu. Deep domain adaptation model with multi-scale residual attention for incipient fault detection of bearings [J]. Journal of Computer Applications, 2020, 40(10): 2890-2898.
[12]	WU Kuan, QIN Pinle, CHAI Rui, ZENG Jianchao. Benign and malignant diagnosis of thyroid nodules based on different ultrasound imaging [J]. Journal of Computer Applications, 2020, 40(1): 77-82.
[13]	LI Minze, LI Xiaoxia, WANG Xueyuan, SUN Wei. Real-time facial expression recognition based on convolutional neural network with multi-scale kernel feature [J]. Journal of Computer Applications, 2019, 39(9): 2568-2574.
[14]	GONG Qin, LEI Man, WANG Jichao, WANG Baoqun. Cross-domain sentiment classification method of convolution-bi-directional long short-term memory based on attention mechanism [J]. Journal of Computer Applications, 2019, 39(8): 2186-2191.
[15]	WANG Ping, PANG Wenhao. Two-stream CNN for action recognition based on video segmentation [J]. Journal of Computer Applications, 2019, 39(7): 2081-2086.