求解多示例问题的支持向量数据描述方法

doi:10.3969/j.issn.0372-2112.2013.04.023

电子学报 ›› 2013, Vol. 41 ›› Issue (4): 763-767.DOI: 10.3969/j.issn.0372-2112.2013.04.023

求解多示例问题的支持向量数据描述方法

方景龙¹, 王万良², 王兴起¹, 龙哲¹, 祁萌¹

1. 杭州电子科技大学图形图像研究所,浙江杭州 310018;
2. 浙江工业大学计算机科学与技术学院,浙江杭州 310023

收稿日期:2012-03-22 修回日期:2012-12-28 出版日期:2013-04-25
- 作者简介:
- 方景龙 男,1964年1月出生于江西景德镇.研究员、博士生导师.1988年和2012年分别在华中理工大学和浙江工业大学获理学硕士和工学博士学位.主要从事机器学习、信息处理和目标探测等方面的研究工作. E-mail:fjl@hdu.edu.cn 王万良 男,1957年6月出生于江苏高邮.教授、博士生导师.2001年在同济大学获工学博士学位.主要从事计算机控制与智能自动化、计算机网络控制系统、智能控制与智能管理分析等方面的研究工作.
- 基金资助:
- 国防基础科研重点项目 (No.A3920110002); 国防技术基础科研项目 (No.Z202012A001)

Support Vector Data Description Method for Solving Multiple Instance Problems

FANG Jing-long¹, WANG Wan-liang², WANG Xing-qi¹, LONG Zhe¹, QI Meng¹

1. Institute of Graphic and Image, Hangzhou Dianzi University, Hangzhou, Zhejiang 310018, China;
2. School of Computer Science and Technology, Zhejiang University of Technology, Hangzhou, Zhejiang 310023, China

Received:2012-03-22 Revised:2012-12-28 Online:2013-04-25 Published:2013-04-25
- Supported by:
- Key Program of National Defense Basic Scientific Research (No.A3920110002); Science and Technology Research Program of Technology Foundation of National Defense (No.Z202012A001)

摘要/Abstract

摘要： 将支持向量数据描述方法引入到多示例学习中,提出了三种基于SVDD的多示例学习方法:基于包分类和示例分类的SVDD多示例学习方法MI-SVDD和mi-SVDD,以及基于正示例预测的SVDD多示例学习方法SVDD-MILD_I.在MUSK数据集上的实验结果表明,MI-SVDD方法和mi-SVDD方法的精度与MI-SVM方法和mi-SVM的总体相当,SVDD-MILD_I方法的精度比较高,超过了我们已知的目前已发表的最好结果;对COREL图像库进行基于内容的图像检索的实验表明,SVDD-MILD_I方法的精度较高,并且比较好地区分了容易混淆的Beach类图像与Mountains类图像.

关键词: 机器学习, 模式分类, 多示例学习, 支持向量数据描述

Abstract: Support Vector Data Description(SVDD)is introduced into multiple instance learning.Three multi-instance learning methods based on SVDD are presented,which include Multiple Instance Learning based on SVDD and bag classification(mi-SVDD)or instance classification(MI-SVDD),and Multiple Instance Learning based on SVDD and positive instance prediction(SVDD-MILD-I).Experimental results on MUSK dataset show that precisions of mi-SVDD and MI-SVDD are quite comparable to those of mi-SVM and MI-SVM;SVDD-MILD-I can get highest accuracy among all the methods known so far.Experimental results in the application of content based image retrieval in COREL image collections demonstrate that precision achieved by SVDD-MILD_I is higher than the others.Additionally,SVDD-MILD_I discriminates the misclassification-prone images between Beach and Mountains quite well.

Key words: machine leaning, pattern classification, multi-instance learning, support vector data description

中图分类号:

TN391.4

方景龙, 王万良, 王兴起, 等. 求解多示例问题的支持向量数据描述方法[J]. 电子学报, 2013, 41(4): 763-767.

FANG Jing-long, WANG Wan-liang, WANG Xing-qi, et al. Support Vector Data Description Method for Solving Multiple Instance Problems[J]. Acta Electronica Sinica, 2013, 41(4): 763-767.

参考文献

[1] Dietterich T G,Lathrop R H,et al.Solving the multiple-instance problem with axi s-parallel rectangles[J].Artificial Intelligence,1997,89(1-2):31-71.
[2] Maron O,Lozano-pérez T.A framework for multiple-instance learning[A].Procee dings of the 1997 Advances in Neural Information Processing Systems Conference [C].Cambridge,MA:MIT Press,1998.570-576.
[3] Andrews S,Hofmann T,et al.Support vector machines for multiple-instance learnin g[A].Proceedings of the 2002 Advances in Neural Information Processing Systems Conference [C].Cambridge,MA:MIT Press,2003.561-568.
[4] Chen Y,Wang J Z.Image categorization by learning and reasoning with regions[J] .Journal of Machine Learning Research,2004,5:913-939.
[5] Chen Y,Bi J,et al.MILES:multiple-instance learning via embedded instance select ion[J],IEEE Transactions on Pattern Analysis and Machine Intelligence,2006,28( 12):1931-1947.
[6] Li Y F,Kwok J T,et al.A convex method for locating regions of interest with mult i-instance learning[A].Proceedings of the European Conference on Machine Lear ning and Principles and Practice of Knowledge Discovery in Databases [C].Bled, Slovenia:Springer,2009.15-30.
[7] Li W J,Yeung D Y.MILD:multiple-instance learning via disambiguation[J].IEEE T ransactions on Knowledge and Data Engineering,2010,22(1):76-89.
[8] Rahmani R,Goldman S A.MISSL:multiple-instance semi-supervised learning[A].Pr oceedings of the 23rd international conference on Machine learning[C].New York :ACM,2006.705-712.
[9] Zhou Z H,Xu J M.On the relation between multi-instance learning and semi-super vised learning[A].Proceedings of the 24th international conference on Machine learning[C].New York:ACM,2007.1167-1174.
[10] Wang H Y,Yang Q,et al.Adaptive p-posterior mixture-model kernels for multiple instance learning[A].Proceedings of the 25th international conference on Machi ne learning[C].New York:ACM,2008.1136-1143.
[11] Zhou Z H,Sun Y Y,et al.Multi-instance learning by treating instances as non-II D samples[A].Proceedings of the 26th Annual International Conference on Machin e Learning[C].New York:ACM,2009.1249-1256.
[12] Deselaers T,Ferrari V.A conditional random field for multiple-instance learning [A].Proceedings of the 27th International Conference on Machine Learning[C].New York:ACM,2010.21-24.
[13] Fu Z,Robles-Kelly A,et al.MILIS:multiple instance learning with instance selec tion[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33 (5):958-977.
[14] 胡文军,王士同,等.适合大样本快速训练的最大夹角间隔核心集向量机[J].电子学报,2 011,39(5):1178-1184. HU Wen-jun,WANG Shi-tong,et al.Maximum vector-angular margin core vector mac hine suitable for fast training for large datasets[J].Acta Electronica Sinica, 2011,39(5):1178-1184.(in Chinese)
[15] 朱孝开,杨德贵.基于推广能力测度的多类SVDD模式识别方法[J].电子学报,2009,37(3 ):464-469. ZHU Xiao-kai,YANG De-gui.Multi-class support vector domain description for pa ttern recognition based on a measure of expansibility[J].Acta Electronica Sini ca,2009,37(3):464-469.(in Chinese)

求解多示例问题的支持向量数据描述方法

Support Vector Data Description Method for Solving Multiple Instance Problems

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	吴军君, 王涛, 王英楷, 王星辉. 基于迁移学习的LiPON制备工艺模拟优化[J]. 电子学报, 2023, 51(3): 687-693.
[2]	李雪莹, 王田路, 梁鹏, 王翀. 基于系统模型的用户评论中非功能需求的自动分类[J]. 电子学报, 2022, 50(9): 2079-2089.
[3]	黄璐, 蔚保国, 李宏生, 李隽, 贾浩男, 程建强, 李雅宁. GNSS拒止环境下的伪卫星指纹定位方法[J]. 电子学报, 2022, 50(4): 811-822.
[4]	崔亚奇, 何友, 唐田田, 熊伟. 一种深度学习航迹关联方法[J]. 电子学报, 2022, 50(3): 759-763.
[5]	蒋华伟, 郭陶, 杨震. 车辆路径问题研究进展[J]. 电子学报, 2022, 50(2): 480-492.
[6]	张笑宇, 沈超, 蔺琛皓, 李前, 王骞, 李琦, 管晓宏. 面向机器学习模型安全的测试与修复[J]. 电子学报, 2022, 50(12): 2884-2918.
[7]	郑锐, 汪秋云, 林卓庞, 靖蓉琦, 姜政伟, 傅建明, 汪姝玮. 一种基于威胁情报层次特征集成的挖矿恶意软件检测方法[J]. 电子学报, 2022, 50(11): 2707-2715.
[8]	张梦, 郑建宏, 刘香燕, 何云. 基于集成学习的全双工中继系统安全中继选择方案研究[J]. 电子学报, 2021, 49(9): 1852-1856.
[9]	郑海颖, 王峰, 姜维, 王志强, 姚西文. 神经网络训练策略对高分辨率遥感图像场景分类性能影响的评估[J]. 电子学报, 2021, 49(8): 1599-1614.
[10]	周东明, 张灿龙, 李志欣, 王智文. 基于多层级视觉融合的图像描述模型[J]. 电子学报, 2021, 49(7): 1286-1290.
[11]	侯秀聪, 赖晓平, 曹九稳. 正则化超限学习机的最大分划广义交替方向乘子法[J]. 电子学报, 2021, 49(4): 625-630.
[12]	陈长青, 郭春, 崔允贺, 申国伟, 蒋朝惠. 基于API短序列的勒索软件早期检测方法[J]. 电子学报, 2021, 49(3): 586-595.
[13]	刘润滋, 吴伟华, 张文柱, 周笛, 张琰. 基于图学习的密集空间网络传输资源调度方法[J]. 电子学报, 2021, 49(11): 2133-2137.
[14]	仇祝令, 查宇飞, 吴敏, 王青. 基于注意力学习的正则化相关滤波跟踪算法[J]. 电子学报, 2020, 48(9): 1762-1768.
[15]	宋宇琦, 高旻, 李骏东, 荣文戈, 熊庆宇. 网络欺凌检测综述[J]. 电子学报, 2020, 48(6): 1220-1229.