08/05/2024, 05:25 PM UTC
超低功耗突触阵列用于神经形态计算Ultra-Low Power Synaptic Arrays for Neuromorphic Computing
➀ 中国研究人员开发了具有持久光电导性的超低功耗碳纳米管/卟啉突触阵列，用于神经形态计算。➁ 这些阵列模仿生物突触，提高了人工神经网络的性能。➂ 这些设备在广泛温度范围内表现出稳定的性能，并在自动驾驶车辆导航任务中显示出高预测准确性。➀ Researchers from China developed ultra-low-power carbon nanotube/porphyrin synaptic arrays with persistent photoconductivity for neuromorphic computing. ➁ These arrays mimic biological synapses, enhancing the performance of artificial neural networks. ➂ The devices showed stable performance across a wide temperature range and high prediction accuracy in autonomous vehicle navigation tasks.
---
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）。
