|
|||||
|
|
| 基于特征联合概率分布和实例的迁移学习算法* | |
| 引用本文： | 赵鹏,吴国琴,刘慧婷,姚晟.基于特征联合概率分布和实例的迁移学习算法*[J].模式识别与人工智能,2016,29(8):717-725. |
| 作者姓名： | 赵鹏 吴国琴 刘慧婷 姚晟 |
| 作者单位： | 安徽大学 计算智能与信号处理教育部重点实验室 合肥 230039 安徽大学 计算机科学与技术学院 合肥 230601 |
| 基金项目： | 国家自然科学基金项目(No.61602004,61472001)、安徽省自然科学基金项目(No.1508085MF127,1408085MF122)、安徽省高校自然科学研究重点项目(No.KJ2016A041)、安徽大学信息保障技术协同创新中心公开招标课题(No.ADXXBZ2014-5,ADXXBZ2014-6)资助 |
| 摘 要： | 针对在单一匹配边缘概率分布以缩减源域和目标域的差异性时存在的泛化能力差的问题，提出联合边缘概率分布和条件概率分布减小域间差异性的基于特征和实例的迁移学习算法.通过核主成分分析在子空间中寻找样本新的特征表示，在该子空间中利用最小化最大均值差异，联合匹配边缘概率分布和条件概率分布以减小源域和目标域间的差异性.同时利用L2，1范数约束选择源域中相关实例进行训练，进一步提高迁移学习获得的模型泛化性能.在字符集和对象识别数据集上的实验表明文中算法的有效性. |
| 关 键 词： | 迁移学习 无监督学习 域自适应 特征映射 |
| 收稿时间： | 2015-12-02 |
Feature Joint Probability Distribution and Instance Based Transfer Learning Algorithm |
|
| ZHAO Peng,WU Guoqin,LIU Huiting,YAO Sheng.Feature Joint Probability Distribution and Instance Based Transfer Learning Algorithm[J].Pattern Recognition and Artificial Intelligence,2016,29(8):717-725. | |
| Authors: | ZHAO Peng WU Guoqin LIU Huiting YAO Sheng |
| Affiliation: | Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education,Anhui University, Hefei 230039 School of Computer Science and Technology, Anhui University, Hefei 230601 |
| Abstract: | Aiming at the poor generalization ability of only matching marginal probability distribution to reduce the domain difference, a feature joint probability distribution and instance based transfer learning algorithm (FJPD-ITLA) is proposed. The instances are represented with the kernel principal component analysis in subspace. In this subspace, the maximum mean discrepancy is expanded to jointly match the marginal and conditional probability distribution. Thus, the difference between the source domain and target domain is reduced. Meanwhile, the L2,1-norm constraint is utilized to choose relevant instances in the source domain, and the generalization ability of the model obtained by transfer learning is improved further. Experimental results on the digital and object recognition datasets demonstrate the validity and efficiency of the proposed algorithm. |
| Keywords: | Transfer Learning Unsupervised Learning Domain Adaption Feature Mapping |
| 点击此处可从《模式识别与人工智能》浏览原始摘要信息 | |
| 点击此处可从《模式识别与人工智能》下载全文 | |