Scholarlink

This chapter focuses on the fundamentals of deep neural networks. The basic concepts of multilayer perceptron are firstly introduced. And then an overview of feature extraction methods regarding images, sequences, and graph structures is provided. Specifically, the image feature extraction plays an important role in environmental visual perception, sequential feature extraction is the key to language understanding and action generation, and graph feature extraction enables the embodied agent to leverage external knowledge and fully exploits their morphological structural properties. Finally, some useful tools for object detection are introduced.

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

Deep Neural Networks Fundamentals

  • 章节

  • 发表日期：

    2026年

查看全文
  • Di Guo,
  • Huaping Liu

摘要

This chapter focuses on the fundamentals of deep neural networks. The basic concepts of multilayer perceptron are firstly introduced. And then an overview of feature extraction methods regarding images, sequences, and graph structures is provided. Specifically, the image feature extraction plays an important role in environmental visual perception, sequential feature extraction is the key to language understanding and action generation, and graph feature extraction enables the embodied agent to leverage external knowledge and fully exploits their morphological structural properties. Finally, some useful tools for object detection are introduced.