Liu Yuqian, Wang Yuechao, Zhang Chi, Yu Liao, Fang Ying, Chen Feng
Department of Automation, Tsinghua University, Beijing 100084, China.
The College of Computer and Cyber Security, Fujian Normal University, Fuzhou 350117, China.
iScience. 2025 Jun 3;28(7):112491. doi: 10.1016/j.isci.2025.112491. eCollection 2025 Jul 18.
Spiking neural networks (SNNs) inspired by biological neurons offer energy-efficient and interpretable computation but is limited by the simplistic structure of point neurons. We introduce a multi-compartment spiking neuron model (MCN) with trainable cross-compartment connections that simulate soma-dendrite interactions. Theoretically, we prove that these connections act as spatiotemporal momentum, guiding learning dynamics toward global optima. To leverage this, we propose a multi-compartment spatiotemporal backpropagation (MCST-BP) algorithm that enhances gradient flow stability. Experimental results for multiple benchmark datasets, including S-MNIST, CIFAR-10, Spiking Heidelberg Digits (SHD), and ECG, show that MC-SNNs outperform traditional SNNs in both convergence speed and accuracy. Our work bridges neurobiological structure and computational modeling, providing a theoretical and practical foundation for high-performance brain-inspired learning systems.
受生物神经元启发的脉冲神经网络（SNN）提供了节能且可解释的计算方式，但受限于点神经元的简单结构。我们引入了一种具有可训练跨隔室连接的多隔室脉冲神经元模型（MCN），该模型模拟了胞体 - 树突相互作用。从理论上讲，我们证明这些连接充当时空动量，引导学习动态朝着全局最优解发展。为了利用这一点，我们提出了一种多隔室时空反向传播（MCST - BP）算法，该算法增强了梯度流稳定性。针对多个基准数据集的实验结果，包括S - MNIST、CIFAR - 10、脉冲海德堡数字（SHD）和心电图（ECG），表明多隔室脉冲神经网络（MC - SNN）在收敛速度和准确性方面均优于传统的脉冲神经网络。我们的工作搭建了神经生物学结构与计算建模之间的桥梁，为高性能脑启发学习系统提供了理论和实践基础。
