计算机应用研究2024，Vol.41Issue(6)：1749-1755,7.DOI:10.19734/j.issn.1001-3695.2023.09.0501
基于自适应重加权和正则化的集成元学习算法
Ensemble meta net based on adaptive reweight and regularization
摘要
Abstract
Deep neural networks tend to overfit to biased training data when there are noisy labels or imbalanced class distri-butions in the training set.Using reweighting strategies with appropriate sample weighting is a common method to address this issue.However,improper reweighting schemes will introduce additional overhead and bias to the network's learning process,it is difficult to solve overfitting problems in biased distribution networks using only reweighting methods.To address this prob-lem,this paper proposed a method that combined label smoothing regularization,class margin regularization,and reweighting,and presented an EMN method based on adaptive reweighting and regularization,which consisted of a base network for classifi-cation and an ensemble meta-net for hyperparameter estimation.The method first obtained the sample loss through the base network,then used three meta-learners to estimate the hyperparameters of adaptive reweighting and regularization in an inte-grated manner based on the loss,and finally used the three hyperparameters to calculate the final ensemble meta-loss and up-date the base network,thereby improving its performance on biased distribution datasets.Experimental results demonstrate that EMN achieves higher accuracy on CIFAR and OCTMNIST datasets compared to other methods,and the effectiveness of diffe-rent strategies are demonstrated through policy correlation analysis.关键词
噪声标签/不平衡/元学习/重加权/正则化Key words
noise label/imbalance/meta learning/reweight/regularization分类
信息技术与安全科学引用本文复制引用
王佳琦,袁野,朱永同,李清都,刘娜..基于自适应重加权和正则化的集成元学习算法[J].计算机应用研究,2024,41(6):1749-1755,7.基金项目
国家自然科学基金资助项目(92048205,61773083) （92048205,61773083）
上海市浦江人才计划资助项目(2019PJD035) （2019PJD035）
