Scholarlink

The recent success of deep neural networks in modeling complex perceptual and cognitive functions has led to comparisons between these artificial networks and the biological neural networks that make up animal and human brains. In particular, the layered architecture of deep neural networks, with each layer specialized for extracting specific types of features from the input, has invited analogies to the modular organization of brain regions specialized for certain cognitive processes. However, modularity theories of brain function overemphasize specialized areas at the expense of distributed processing and interconnectivity.

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

The Parallels Between Deep Neural Networks and Modularity Theories of Brain Function

  • 章节

  • 发表日期：

    2025年

查看全文
  • Paolo Massimo Buscema,
  • Weldon A. Lodwick,
  • Giulia Massini,
  • Pier Luigi Sacco,
  • Masoud Asadi-Zeydabadi,
  • Francis Newman,
  • Riccardo Petritoli,
  • Marco Breda

摘要

The recent success of deep neural networks in modeling complex perceptual and cognitive functions has led to comparisons between these artificial networks and the biological neural networks that make up animal and human brains. In particular, the layered architecture of deep neural networks, with each layer specialized for extracting specific types of features from the input, has invited analogies to the modular organization of brain regions specialized for certain cognitive processes. However, modularity theories of brain function overemphasize specialized areas at the expense of distributed processing and interconnectivity.