Zhang Xumeng, Lu Jian, Wang Zhongrui, Wang Rui, Wei Jinsong, Shi Tuo, Dou Chunmeng, Wu Zuheng, Zhu Jiaxue, Shang Dashan, Xing Guozhong, Chan Mansun, Liu Qi, Liu Ming
Frontier Institute of Chip and System, Fudan University, Shanghai 200433, China; Key Laboratory of Microelectronic Devices & Integrated Technology, Institute of Microelectronics, Chinese Academy of Sciences, Beijing 100029, China; University of Chinese Academy of Sciences, Beijing 100049, China.
Key Laboratory of Microelectronic Devices & Integrated Technology, Institute of Microelectronics, Chinese Academy of Sciences, Beijing 100029, China.
Sci Bull (Beijing). 2021 Aug 30;66(16):1624-1633. doi: 10.1016/j.scib.2021.04.014. Epub 2021 Apr 17.
Spiking neural network, inspired by the human brain, consisting of spiking neurons and plastic synapses, is a promising solution for highly efficient data processing in neuromorphic computing. Recently, memristor-based neurons and synapses are becoming intriguing candidates to build spiking neural networks in hardware, owing to the close resemblance between their device dynamics and the biological counterparts. However, the functionalities of memristor-based neurons are currently very limited, and a hardware demonstration of fully memristor-based spiking neural networks supporting in-situ learning is very challenging. Here, a hybrid spiking neuron combining a memristor with simple digital circuits is designed and implemented in hardware to enhance neuron functions. The hybrid neuron with memristive dynamics not only realizes the basic leaky integrate-and-fire neuron function but also enables the in-situ tuning of the connected synaptic weights. Finally, a fully hardware spiking neural network with the hybrid neurons and memristive synapses is experimentally demonstrated for the first time, and in-situ Hebbian learning is achieved with this network. This work opens up a way towards the implementation of spiking neurons, supporting in-situ learning for future neuromorphic computing systems.
脉冲神经网络受人类大脑启发，由脉冲神经元和可塑性突触组成，是神经形态计算中高效数据处理的一种有前途的解决方案。最近，基于忆阻器的神经元和突触因其器件动力学与生物对应物之间的高度相似性，正成为构建硬件脉冲神经网络的有趣候选者。然而，基于忆阻器的神经元的功能目前非常有限，并且支持原位学习的完全基于忆阻器的脉冲神经网络的硬件演示极具挑战性。在此，设计并在硬件中实现了一种将忆阻器与简单数字电路相结合的混合脉冲神经元，以增强神经元功能。具有忆阻动力学的混合神经元不仅实现了基本的泄漏积分发放神经元功能，还能对连接的突触权重进行原位调整。最后，首次通过实验展示了一个具有混合神经元和忆阻突触的全硬件脉冲神经网络，并利用该网络实现了原位赫布学习。这项工作为实现支持未来神经形态计算系统原位学习的脉冲神经元开辟了一条道路。
