带有错误标签的张量数据的稳健多分类模型  

Robust Multiclass Models for Mislabeled Tensor Data

作　　者：张家瑞 樊亚莉[1] Jiarui Zhang;Yali Fan(College of Science,University of Shanghai for Science and Technology,Shanghai)

机构地区：[1]上海理工大学理学院,上海

出　　处：《运筹与模糊学》2024年第3期242-255,共14页Operations Research and Fuzziology

基　　金：国家自然科学基金资助项目(12371308)。

摘　　要：传统机器学习方法大多都是基于正确标签的训练数据进行监督学习,但实际观测到的训练数据标签极可能受到污染,而错误标签的存在会导致传统模型产生有偏估计。现存的关于错误标签的稳健模型往往基于向量数据进行分类,面对存在错误标签的高阶张量数据时只能将其转化为低阶格式,由此产生过拟合问题且破坏张量结构。针对上述问题提出一种稳健的张量多分类模型(RMLTMLR),基于最小γ-散度估计、张量管道秩及相应的核范数来处理带有错误标签的低秩张量,在利用张量结构特点的同时使模型对污染标签具有稳健性,提高多分类准确率。进行的大量实验表明RMLTMLR模型在不同类别和污染程度的张量数据上有着优良的分类效果,与非稳健的模型相比,分类准确率显著提升。Most of the traditional machine learning methods perform supervised learning based on training data with correct labels.However,the actual observed training data labels are likely to be conta-minated,and the existence of wrong labels will lead to biased estimates of the traditional model.The existing robust models for mislabel classification are often based on vector data.When facing high-order tensor data with mislabels,they have to transform it into low-order format,resulting in overfitting problem and damage to the tensor structure.Aiming at the above problems,a robust tensor multi-classification model(RMLTMLR)is proposed,which is based on minimumγ-divergence estimation,tensor tubal rank and the corresponding nuclear norm to deal with low-rank tensors with wrong labels.The model is robust to contaminated labels while taking advantage of the structural characteristics of tensors,and improves the accuracy of multi-classification.A large number of experiments show that the RMLTMLR model has excellent classification effects on ten-sor data with different categories and pollution levels,and the classification accuracy is signifi-cantly improved compared with the non-robust model.

关 键 词：图像多分类 错误标签 低秩张量 张量管道秩 机器学习 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]