Scholarlink

<p>Graph data is crucial for modeling complex relationships in various fields, but conventional graph computing methods struggle to handle increasingly intricate and large-scale graph data. Electric current-based graph computing and Quantum-inspired graph computing offer innovative hardware-based solutions to these challenges. Electric current-based graph computing has progressed from Euclidean graph data to non-Euclidean ones using the memristive crossbar arrays. This Perspective introduces various crossbar array-based electric current-based graph computings, which offer flexibility in representing complex graphs, enabling a wide range of graphical applications in materials, biology, and social science. It also discusses quantum-inspired graph computing, employing probabilistic bits, oscillatory neural networks, and related architectures to solve complex optimization problems. Electric current-based and quantum-inspired graph computing remain in their early stages of evolution, requiring further work to advance materials, devices, and architectures to fully realize their potential. These advancements will open opportunities for more diverse and complex real-world applications.</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

Next-generation graph computing with electric current-based and quantum-inspired approaches

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年8月28日

查看全文
  • Yoon Ho Jang,
  • Janguk Han,
  • Soo Hyung Lee,
  • Cheol Seong Hwang

摘要

Graph data is crucial for modeling complex relationships in various fields, but conventional graph computing methods struggle to handle increasingly intricate and large-scale graph data. Electric current-based graph computing and Quantum-inspired graph computing offer innovative hardware-based solutions to these challenges. Electric current-based graph computing has progressed from Euclidean graph data to non-Euclidean ones using the memristive crossbar arrays. This Perspective introduces various crossbar array-based electric current-based graph computings, which offer flexibility in representing complex graphs, enabling a wide range of graphical applications in materials, biology, and social science. It also discusses quantum-inspired graph computing, employing probabilistic bits, oscillatory neural networks, and related architectures to solve complex optimization problems. Electric current-based and quantum-inspired graph computing remain in their early stages of evolution, requiring further work to advance materials, devices, and architectures to fully realize their potential. These advancements will open opportunities for more diverse and complex real-world applications.