Lin Xianghong, Zhang Zhen, Zheng Donghao
College of Computer Science and Engineering, Northwest Normal University, Lanzhou 730070, China.
Brain Sci. 2023 Jan 18;13(2):168. doi: 10.3390/brainsci13020168.
By mimicking the hierarchical structure of human brain, deep spiking neural networks (DSNNs) can extract features from a lower level to a higher level gradually, and improve the performance for the processing of spatio-temporal information. Due to the complex hierarchical structure and implicit nonlinear mechanism, the formulation of spike train level supervised learning methods for DSNNs remains an important problem in this research area. Based on the definition of kernel function and spike trains inner product (STIP) as well as the idea of error backpropagation (BP), this paper firstly proposes a deep supervised learning algorithm for DSNNs named BP-STIP. Furthermore, in order to alleviate the intrinsic weight transport problem of the BP mechanism, feedback alignment (FA) and broadcast alignment (BA) mechanisms are utilized to optimize the error feedback mode of BP-STIP, and two deep supervised learning algorithms named FA-STIP and BA-STIP are also proposed. In the experiments, the effectiveness of the proposed three DSNN algorithms is verified on the MNIST digital image benchmark dataset, and the influence of different kernel functions on the learning performance of DSNNs with different network scales is analyzed. Experimental results show that the FA-STIP and BP-STIP algorithms can achieve 94.73% and 95.65% classification accuracy, which apparently possess better learning performance and stability compared with the benchmark algorithm BP-STIP.
通过模仿人类大脑的层次结构，深度脉冲神经网络（DSNNs）可以逐步从低层次到高层次提取特征，并提高处理时空信息的性能。由于其复杂的层次结构和隐含的非线性机制，为DSNNs制定脉冲序列级监督学习方法仍然是该研究领域的一个重要问题。基于核函数和脉冲序列内积（STIP）的定义以及误差反向传播（BP）的思想，本文首先提出了一种用于DSNNs的深度监督学习算法，名为BP-STIP。此外，为了缓解BP机制固有的权重传输问题，利用反馈对齐（FA）和广播对齐（BA）机制来优化BP-STIP的误差反馈模式，并提出了两种深度监督学习算法，名为FA-STIP和BA-STIP。在实验中，在MNIST数字图像基准数据集上验证了所提出的三种DSNN算法的有效性，并分析了不同核函数对不同网络规模的DSNNs学习性能的影响。实验结果表明，FA-STIP和BP-STIP算法可以分别达到94.73%和95.65%的分类准确率，与基准算法BP-STIP相比，它们显然具有更好的学习性能和稳定性。
