典型文献
Data fusing and joint training for learning with noisy labels
文献摘要：
It is well known that deep learning depends on a large amount of clean data.Because of high annotation cost,various methods have been devoted to annotating the data automatically.However,a larger number of the noisy labels are generated in the datasets,which is a challenging problem.In this paper,we propose a new method for selecting training data accurately.Specifically,our approach fits a mixture model to the per-sample loss of the raw label and the predicted label,and the mixture model is utilized to dynamically divide the training set into a correctly labeled set,a correctly predicted set,and a wrong set.Then,a network is trained with these sets in the supervised learning manner.Due to the confirmation bias problem,we train the two networks alternately,and each network establishes the data division to teach the other network.When optimizing network parameters,the labels of the samples fuse respectively by the probabilities from the mixture model.Experiments on CIFAR-10,CIFAR-100 and Clothing1M demonstrate that this method is the same or superior to the state-of-the-art methods.
文献关键词：
中图分类号：
作者姓名：
Yi WEI;Mei XUE;Xin LIU;Pengxiang XU
作者机构：
College of Electrical Engineering and Control Science,Nanjing Tech University,Nanjing 211816,China;Beijing Seetatech Technology Co.,Ltd,Beijing 100029,China
文献出处：
引用格式：
[1]Yi WEI;Mei XUE;Xin LIU;Pengxiang XU-.Data fusing and joint training for learning with noisy labels)[J].计算机科学前沿,2022(06):59-66
A类：
annotating,Clothing1M
B类：
Data,fusing,joint,training,learning,noisy,labels,It,well,known,that,deep,depends,amount,clean,Because,high,annotation,cost,various,methods,have,been,devoted,automatically,However,larger,number,are,generated,datasets,which,challenging,problem,In,this,paper,propose,new,selecting,accurately,Specifically,our,approach,fits,mixture,model,loss,raw,predicted,utilized,dynamically,divide,into,correctly,labeled,wrong,Then,trained,these,supervised,manner,Due,confirmation,bias,networks,alternately,establishes,division,teach,other,When,optimizing,parameters,samples,fuse,respectively,by,probabilities,from,Experiments,CIFAR,demonstrate,same,superior,state,art
AB值：
0.562122
相似文献
机标中图分类号，由域田数据科技根据网络公开资料自动分析生成，仅供学习研究参考。
