文档介绍：Advances in Applied Mathematics 应用数学进展, 2022, 11(3), 1275-1281
Published Online March 2022 in Hans. , the Group Lasso regular terms can eliminate
unnecessary weights at the group level and have a good sparse effect. As is known to all, there are
two learning methods for weight update using gradient method: one is batch learning algorithm;
the other is online learning algorithm. This paper proposes an online gradient learning algorithm
with Group Lasso regularized terms to train Pi-Sigma neural networks. Finally, numerical results
show that the improved algorithm converges faster and has better generalization performance.

Keywords
Pi-Sigma Neural Network, Online Gradient Algorithm, Group Lasso Regular Term

Copyright © 2022 by author(s) and Hans Publishers Inc.
This work is licensed under the Creative Commons Attribution International License (CC BY ).
/
Open Access

1. 引言
人工神经网络已经被广泛应用于各种领域[1]。其中前馈神经网络以其结构灵活性，具有与不同训练
算法的兼容性的特点成为最受欢迎的体系结构之一。Pi-Sigma 神经网络(简称 PSNN)由 Shin 和 Ghosh 在
1991 年首先提出，它是前馈神经网络中的一种，具有良好的收敛性和泛化性能，该网络也具有较为规则
的结构，并且具有更快的学****速度，能够通过增量添加单元来