信息与控制2016，Vol.45Issue(4)：463-470,478,9.DOI:10.13976/j.cnki.xk.2016.0463
基于最大均值差异的多标记迁移学习算法
Multi-label Transfer Learning via Maximum Mean Discrepancy
摘要
Abstract
Due to the different distribution of features between the source and target domains in a multi-label trans-fer learning problem,source domain data cannot exert any effect.To resolve this problem,here we propose novel multi-label transfer learning via the maximum mean discrepancy.The proposed algorithm decomposes a relational matrix to learn a common subspace.Furthermore,we incorporate the empirical maximum mean dis-crepancy into the objective function of matrix factorization to minimize the probability distance between differ-ent domains.Experimental results from multi-label classification demonstrate that the proposed approach a-chieves better performance than other similar algorithms in terms of accuracy and efficiency.关键词
多标记/迁移学习/最大均值差异/共享子空间Key words
multi-label/transfer learning/maximum mean discrepancy/shared subspace分类
信息技术与安全科学引用本文复制引用
姜海燕,刘昊天,舒欣,徐彦,伍艳莲,郭小清..基于最大均值差异的多标记迁移学习算法[J].信息与控制,2016,45(4):463-470,478,9.基金项目
国家自然科学基金资助项目（30971697，61403205）；国家863计划资助项目（2013AA100404）；江苏省农业科技自主创新资金（CX（16）1039） （）
