咨询与建议

限定检索结果

文献类型

  • 1 篇 期刊文献

馆藏范围

  • 1 篇 电子文献
  • 0 种 纸本馆藏

日期分布

学科分类号

  • 1 篇 理学
    • 1 篇 数学

主题

  • 1 篇 directed acyclic...
  • 1 篇 autoencoders
  • 1 篇 backpropagation ...

机构

  • 1 篇 department of el...
  • 1 篇 department of el...
  • 1 篇 department of ma...

作者

  • 1 篇 dan schonfeld
  • 1 篇 chirag agarwal
  • 1 篇 joe klobusicky

语言

  • 1 篇 英文
检索条件"作者=Joe Klobusicky"
1 条 记 录,以下是1-10 订阅
排序:
CONVERGENCE OF BACKPRIOPAG ATION WITH MOMENTUM FOR NETWORK A RCHITECTURES WITH SKIP CONNECTIONS
收藏 引用
Journal of Computational Mathematics 2021年 第1期39卷 147-158页
作者: Chirag Agarwal joe klobusicky Dan Schonfeld Department of Electrical and Cormputer Engineering University of Illinois at ChicagoChicagoIL 60607USA Department of Mathematical Sciences Rensselaer Polytechnic InstituteTroyNY 12180USA Department of Electrical and Computer Engineering University of Illinois at ChicagoChicagoIL 60607USA
We study a class of deep neural networks with architectures that form a directed acyclic graph(DAG).For backpropagation defined by gradient descent with adaptive momentum,we show weights converge for a large class of ... 详细信息
来源: 维普期刊数据库 维普期刊数据库 评论