Scholarlink

<p>Fully Connected Network (FCN) layers are widely used in neural networks. Indeed, these layers are proving particularly effective for tasks such as classification and regression. In this paper, we introduce a novel fully connected network architecture called Tensor Einstein Reduced Fully Connected Network (TER-FCN), based on the Einstein product, Tensor Gradient Descent (TGD) algorithm, and Einstein Singular Value Decomposition (E-SVD). In addition, we propose the application of this method in Region-based Convolutional Neural Networks (R-CNN) for object detection. Numerical tests have been carried out to validate the effectiveness of this approach.</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

TER-FCN: Tensor Einstein Reduced Fully Connected Network applied to Region-based Convolutional Neural Networks for object detection

  • 期刊论文

  • 发表日期：

    2025年5月27日

查看全文
  • Alaa EL ICHI,
  • Wissam KADDAH,
  • Marwa EL BOUZ,
  • Isabelle BADOC

摘要

Fully Connected Network (FCN) layers are widely used in neural networks. Indeed, these layers are proving particularly effective for tasks such as classification and regression. In this paper, we introduce a novel fully connected network architecture called Tensor Einstein Reduced Fully Connected Network (TER-FCN), based on the Einstein product, Tensor Gradient Descent (TGD) algorithm, and Einstein Singular Value Decomposition (E-SVD). In addition, we propose the application of this method in Region-based Convolutional Neural Networks (R-CNN) for object detection. Numerical tests have been carried out to validate the effectiveness of this approach.