摘要
偏标记学习是一类重要的弱监督机器学习框架.在该框架下,每个对象在输入空间由单个示例进行刻画,在输出空间与一组候选标记相对应,其中仅有一个标记为其真实标记.利用有歧义性的样本进行建模,直观的策略是对候选标记集合进行消歧,然而该策略会受到伪标记的影响,因此有必要考虑从非消歧的角度解决偏标记学习问题.本文将围绕基于消歧、非消歧策略的偏标记学习算法对该领域进行综述.首先,给出偏标记学习的定义以及其与其他相关学习框架的关系.然后对现有几种代表性基于消歧策略的偏标记学习算法进行介绍.接下来重点介绍我们提出的两种基于非消歧策略的偏标记学习算法.最后对本文进行总结并简要讨论进一步的研究方向.
Partial label learning is an important weakly supervised machine learning framework.In partial label learning,each object is described by a single instance in the input space;however,in the output space,it is associated with a set of candidate labels among which only one is valid.An intuitive strategy is to disambiguate candidate labels,but this strategy tends to be misled by false positive labels;therefore,new disambiguationfree approaches need to be considered.In this paper,several algorithms are reviewed from the perspective of disambiguation and disambiguation-free strategies.First,the problem definition on partial label learning and its relationship with other related learning frameworks are given.Second,several representative partial label learning algorithms via the disambiguation strategy are introduced.Third,two of our proposed disambiguationfree algorithms are presented.Finally,the summary of this paper is given and possible future investigations on partial label learning are briefly discussed.
作者
张敏灵
吴璇
Min-Ling ZHANG;Xuan WU(School of Computer Science and Engineering,Southeast University,Nanjing 210096,China;Key Laboratory of Computer Network and Information Integration(Southeast University),Ministry of Educa-tion,Nanjing 210096,China)
出处
《中国科学：信息科学》
CSCD
北大核心
2019年第9期1083-1096,共14页
Scientia Sinica(Informationis)
基金
国家重点研发计划(批准号:2018YFB1004300)
国家自然科学基金(批准号:61573104)资助项目
关键词
机器学习
弱监督学习
偏标记学习
候选标记
非消歧策略
machine learning
weakly-supervised learning
partial label learning
candidate label
disambiguation-free strategy
