智能系统学报2025，Vol.20Issue(2)：465-474,10.DOI:10.11992/tis.202402018
用于高维小样本特征选择的超网络设计
Hypernetwork design for feature selection of high-dimensional small samples
摘要
Abstract
Feature selection is a widely recognized challenge across various industries.They typically target high-dimen-sional datasets with fewer samples,such as those in biology and medicine field.Many regularization networks outper-form complex network structures on such datasets.However,numerous underlying feature relationships can still be overfitted,particularly with limited data.This study proposes an end-to-end sparse reconstruction network to address this issue.First,the model enhances features through sparsity and singular value embedding.Then,it trains the embed-ding matrix through a parallel auxiliary network to reconstruct prediction weights,which implements a parameter-redu-cing super-network learning approach.This approach reduces the impact of overfitting on networks with fewer paramet-ers,which effectively mitigates the influence of ineffective parameters on the network.Experiments conducted on 12 high-dimensional small-sample datasets in biology and medicine field reveal an average improvement of 3.26 percent-age point in classification accuracy after dimensionality reduction in eight feature selection networks.Furthermore,the roles of the disintegration layer,reconstruction,and correlation layer are separately validated through ablation experi-ments,followed by weight result analysis,which further elucidates the extended applications of the model.关键词
特征选择/正则化网络/过拟合/端到端/稀疏重构/奇异值/辅助网络/超网络/高维小样本Key words
feature selection/regularization network/overfitting/end-to-end/sparse reconstruction/singular value/aux-iliary network/hypernetwork/high-dimensional small sample分类
计算机与自动化引用本文复制引用
魏俊伊,董红斌,余紫康..用于高维小样本特征选择的超网络设计[J].智能系统学报,2025,20(2):465-474,10.基金项目
黑龙江自然科学基金项目(LH2020F023). （LH2020F023）
