电子学报 ›› 2014, Vol. 42 ›› Issue (3): 556-560.DOI: 10.3969/j.iss.0372-2012-2014.03.020

• 科研通信 • 上一篇    下一篇

具有协同约束的共生迁移学习算法研究

张景祥1,2, 王士同1, 邓赵红1, 李奕1, 蒋亦樟1   

  1. 1. 江南大学数字媒体学院, 江苏无锡 214122;
    2. 江南大学理学院, 江苏无锡 214122
  • 收稿日期:2012-11-14 修回日期:2013-01-07 出版日期:2014-03-25
    • 作者简介:
    • 张景祥 男,1977年生于吉林通化.2002年、2007年分别在重庆大学、江南大学获得理学学士、工学硕士学位,2011年进入江南大学攻读博士学位,主要从事模式识别、人工智能和生物信息学等方面的研究.E-mail:zjx145@163.com
    • 基金资助:
    • 国家自然科学基金 (No.61170122,No.61202311,No.61272210); 江苏省自然科学基金 (No.BK2012552)

Symbiosis Transfer Learning Method with Collaborative Constraints

ZHANG Jing-xiang1,2, WANG Shi-tong1, DENG Zhao-hong1, LI Yi1, JIANG Yi-zhang1   

  1. 1. School of Digital Media, Jiangnan University, Wuxi, Jiangsu 214122, China;
    2. School of Science, Jiangnan University, Wuxi, Jiangsu 214122, China
  • Received:2012-11-14 Revised:2013-01-07 Online:2014-03-25 Published:2014-03-25
    • Supported by:
    • National Natural Science Foundation of China (No.61170122, No.61202311, No.61272210); Natural Science Foundation of Jiangsu Province,  China (No.BK2012552)

摘要: 传统迁移学习方法通常直接利用相关领域中的数据来辅助完成当前领域的学习任务,而忽略了领域间互相学习的能力.针对此类问题,提出了一种具有协同约束的共生迁移学习方法(Collaborative Constraints based Symbiosis Transfer Learning,CCSTL).在协同约束的基础上,引入共生迁移机制实现领域间的交替互动学习,进而实现源领域和目标领域间的知识迁移,从而提高受训分类器的分类性能.在模拟数据和真实数据集上的实验结果表明了新算法的有效性.

关键词: 协同约束, 共生迁移学习, 分类, 支持向量机

Abstract: Transfer learning algorithms usually focus on reusing data of related domains to help solving the learning tasks in the target domain.However,these methods ignore the ability of mutual learning between domains.In this paper,a collaborative constraint based symbiosis transfer learning method (CCSTL) is proposed.Symbiotic transfer mechanism is used to implement mutual learning among domains along with the collaborative constraint.With the help of the iterative optimizations,the proposed method can realize knowledge transfer between the source and target domains.Experimental results on synthetic and real world datasets show the superior or comparable performance of the proposed algorithm compared with existing algorithms.

Key words: collaborative constraints, symbiosis transfer learning, classification, support vector machine

中图分类号: