Learning Multi Labels from Single Label——An Extreme Weak Label Learning Algorithm  

作　　者：DUAN Junhong LI Xiaoyu MU Dejun 

机构地区：[1]Shenzhen Research Institute, Northwestern Polytechnical University

出　　处：《Wuhan University Journal of Natural Sciences》2019年第2期161-168,共8页武汉大学学报（自然科学英文版）

基　　金：Supported by the National Natural Science Foundation of China(61672433);the Fundamental Research Fund for Shenzhen Science and Technology Innovation Committee(201703063000511,201703063000517);the National Cryptography Development Fund(MMJJ20170210);the Science and Technology Project of State Grid Corporation of China(522722180007)

摘　　要：This paper presents a novel algorithm for an extreme form of weak label learning, in which only one of all relevant labels is given for each training sample. Using genetic algorithm, all of the labels in the training set are optimally divided into several non-overlapping groups to maximize the label distinguishability in every group. Multiple classifiers are trained separately and ensembled for label predictions. Experimental results show significant improvement over previous weak label learning algorithms.This paper presents a novel algorithm for an extreme form of weak label learning, in which only one of all relevant labels is given for each training sample. Using genetic algorithm, all of the labels in the training set are optimally divided into several non-overlapping groups to maximize the label distinguishability in every group. Multiple classifiers are trained separately and ensembled for label predictions. Experimental results show significant improvement over previous weak label learning algorithms.

关 键 词：weak-supervised LEARNING genetic algorithm MULTI-LABEL classification 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]