12/11/2024, 03:09 PM UTC
用于印刷电子的微型机器学习技术Tiny Machine Learning For Printed Electronics
来自卡尔斯鲁厄理工学院和帕特拉斯大学的学者们开发了一种用于印刷电子的微型机器学习（tinyML）的新方法，旨在优化机器学习模型，使其适用于超低功耗应用。这种方法旨在为可穿戴设备、植入物和其他应用提供高效的神经网络，同时解决传统基于硅的系统局限性。该团队提出的解决方案引入了用于多感官应用的“顺序超级微型ML多层感知器电路”，与现有方法相比，在面积效率和功耗方面取得了显著的改进。Researchers from the Karlsruhe Institute of Technology and the University of Patras have developed a new approach to tiny machine learning (tinyML) for printed electronics, optimizing machine learning models for ultra-low-power applications. This approach aims to enable efficient neural networks for wearables, implants, and other applications, while addressing the limitations of traditional silicon-based systems. The team's solution introduces 'sequential super-tinyML multi-layer perceptron circuits' for multi-sensory applications, achieving significant improvements in area efficiency and power consumption compared to existing methods.
---
本文由大语言模型（LLM）生成，旨在为读者提供半导体新闻内容的知识扩展（Beta）。
