Superiority of quadratic over conventional neural networks for classification of gaussian mixture data

作     者：Tianrui Qi Ge Wang 

作者机构：Department of Computer ScienceRensselaer Polytechnic InstituteNY 12180TroyUSA 

出 版 物：《Visual Computing for Industry,Biomedicine,and Art》 (工医艺的可视计算（英文）)

年 卷 期：2022年第5卷第1期

页      面：279-289页

核心收录：

学科分类：081203[工学-计算机应用技术] 08[工学] 0835[工学-软件工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：This work was supported in part by NIH Nos.R01CA237267 R01HL151561 R21CA264772 and R01EB032716 

主　　题：Artificial neural networks Quadratic neurons Quadratic neural networks Backpropagation Classification Gaussian mixture models 

摘      要：To enrich the diversity of artificial neurons,a type of quadratic neurons was proposed previously,where the inner product of inputs and weights is replaced by a quadratic *** this paper,we demonstrate the superiority of such quadratic neurons over conventional *** this purpose,we train such quadratic neural networks using an adapted backpropagation algorithm and perform a systematic comparison between quadratic and conventional neural networks for classificaiton of Gaussian mixture data,which is one of the most important machine learning *** results show that quadratic neural networks enjoy remarkably better efficacy and efficiency than conventional neural networks in this context,and potentially extendable to other relevant applications.