Zhou Peng, Edwards Alexander J, Mancoff Frederick B, Aggarwal Sanjeev, Heinrich-Barna Stephen K, Friedman Joseph S
Department of Electrical and Computer Engineering, The University of Texas at Dallas, Richardson, TX, USA.
Everspin Technologies Inc., Chandler, AZ, USA.
Commun Eng. 2025 Aug 4;4(1):142. doi: 10.1038/s44172-025-00479-2.
Neuromorphic computing aims to mimic both the function and structure of biological neural networks to provide artificial intelligence with extreme efficiency. Conventional approaches store synaptic weights in non-volatile memory devices with analog resistance states, permitting in-memory computation of neural network operations while avoiding the costs of transferring synaptic weights from memory. However, the use of analog resistance states for storing weights in neuromorphic systems is impeded by stochastic writing, weights drifting over time through stochastic processes, and limited endurance that reduces the precision of synapse weights. Here we propose and experimentally demonstrate neuromorphic networks that provide high-accuracy inference thanks to the binary resistance states of magnetic tunnel junctions (MTJs), while leveraging the analog nature of their stochastic spin-transfer torque (STT) switching for unsupervised Hebbian learning. We performed an experimental demonstration of a neuromorphic network directly implemented with MTJ synapses, for both inference and spike-timing-dependent plasticity learning. We also demonstrated through simulation that the proposed system for unsupervised Hebbian learning with stochastic STT-MTJ synapses can achieve competitive accuracies for MNIST handwritten digit recognition. By appropriately applying neuromorphic principles through hardware-aware design, the proposed STT-MTJ neuromorphic learning networks provide a pathway toward artificial intelligence hardware that learns autonomously with extreme efficiency.
神经形态计算旨在模仿生物神经网络的功能和结构,为人工智能提供极高的效率。传统方法将突触权重存储在具有模拟电阻状态的非易失性存储设备中,允许在内存中进行神经网络运算,同时避免从内存中传输突触权重的成本。然而,在神经形态系统中使用模拟电阻状态来存储权重受到随机写入、权重随时间通过随机过程漂移以及有限耐久性的阻碍,这些因素会降低突触权重的精度。在此,我们提出并通过实验证明了神经形态网络,由于磁隧道结(MTJ)的二进制电阻状态,该网络可提供高精度推理,同时利用其随机自旋转移力矩(STT)开关的模拟特性进行无监督赫布学习。我们对直接用MTJ突触实现的神经形态网络进行了推理和基于脉冲时间的可塑性学习的实验演示。我们还通过模拟证明,所提出的具有随机STT-MTJ突触的无监督赫布学习系统在MNIST手写数字识别中可实现具有竞争力的准确率。通过硬件感知设计适当地应用神经形态原理,所提出的STT-MTJ神经形态学习网络为实现高效自主学习的人工智能硬件提供了一条途径。