Lu Sen, Sengupta Abhronil
School of Electrical Engineering and Computer Science, The Pennsylvania State University, University Park, PA, United States.
Front Neurosci. 2022 Apr 25;16:838523. doi: 10.3389/fnins.2022.838523. eCollection 2022.
Neuromorphic computing algorithms based on Spiking Neural Networks (SNNs) are evolving to be a disruptive technology driving machine learning research. The overarching goal of this work is to develop a structured algorithmic framework for SNN training that optimizes unique SNN-specific properties like neuron spiking threshold using neuroevolution as a feedback strategy. We provide extensive results for this hybrid bio-inspired training strategy and show that such a feedback-based learning approach leads to explainable neuromorphic systems that adapt to the specific underlying application. Our analysis reveals 53.8, 28.8, and 28.2% latency improvement for the neuroevolution-based SNN training strategy on CIFAR-10, CIFAR-100, and ImageNet datasets, respectively in contrast to state-of-the-art conversion based approaches. The proposed algorithm can be easily extended to other application domains like image classification in presence of adversarial attacks where 43.2 and 27.9% latency improvements were observed on CIFAR-10 and CIFAR-100 datasets, respectively.
基于脉冲神经网络（SNN）的神经形态计算算法正在演变成一种推动机器学习研究的颠覆性技术。这项工作的总体目标是为SNN训练开发一个结构化算法框架，该框架使用神经进化作为反馈策略来优化独特的SNN特定属性，如神经元脉冲阈值。我们为这种混合生物启发式训练策略提供了大量结果，并表明这种基于反馈的学习方法会产生适用于特定基础应用的可解释神经形态系统。我们的分析表明，与基于最先进转换的方法相比，基于神经进化的SNN训练策略在CIFAR-10、CIFAR-100和ImageNet数据集上的延迟分别提高了53.8%、28.8%和28.2%。所提出的算法可以很容易地扩展到其他应用领域，如存在对抗性攻击时的图像分类，在CIFAR-10和CIFAR-100数据集上分别观察到延迟提高了43.2%和27.9%。
