Scholarlink

<p>Canonical correlation analysis is a classic statistical technique. In this paper, we develop a neural network method based on the Riemannian gradient for solving canonical correlation analysis problems. For the theoretical analysis, the geometric dynamics properties of this method are investigated. Numerical experiments indicate the feasibility and effectiveness of the proposed neural network method.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Riemannian-based neural network method for solving canonical correlation analysis

  • 期刊论文

  • 发表日期：

    2025年4月11日

查看全文
  • Zhuo-Cheng Xie,
  • Ming Wang,
  • Yu-Hang Wang,
  • Huan Ren

摘要

Canonical correlation analysis is a classic statistical technique. In this paper, we develop a neural network method based on the Riemannian gradient for solving canonical correlation analysis problems. For the theoretical analysis, the geometric dynamics properties of this method are investigated. Numerical experiments indicate the feasibility and effectiveness of the proposed neural network method.