Habara Takehiro, Sato Takashi, Awano Hiromitsu
Department of Communications and Computer Engineering, Graduate School of Informatics, Kyoto University, Kyoto, Japan.
Front Neurosci. 2024 Aug 5;18:1420119. doi: 10.3389/fnins.2024.1420119. eCollection 2024.
Spiking neural networks (SNNs) have garnered significant attention due to their notable energy efficiency. However, conventional SNNs rely on spike firing frequency to encode information, necessitating a fixed sampling time and leaving room for further optimization. This study presents a novel approach to reduce sampling time and conserve energy by extracting early prediction results from the intermediate layer of the network and integrating them with the final layer's predictions in a Bayesian fashion. Experimental evaluations conducted on image classification tasks using MNIST, CIFAR-10, and CIFAR-100 datasets demonstrate the efficacy of our proposed method when applied to VGGNets and ResNets models. Results indicate a substantial energy reduction of 38.8% in VGGNets and 48.0% in ResNets, illustrating the potential for achieving significant efficiency gains in spiking neural networks. These findings contribute to the ongoing research in enhancing the performance of SNNs, facilitating their deployment in resource-constrained environments. Our code is available on GitHub: https://github.com/hanebarla/BayesianSpikeFusion.
脉冲神经网络（SNNs）因其显著的能源效率而备受关注。然而，传统的脉冲神经网络依靠脉冲发放频率来编码信息，这需要固定的采样时间，仍有进一步优化的空间。本研究提出了一种新方法，通过从网络中间层提取早期预测结果，并以贝叶斯方式将其与最后一层的预测结果相结合，以减少采样时间并节约能源。使用MNIST、CIFAR-10和CIFAR-100数据集对图像分类任务进行的实验评估表明，我们提出的方法应用于VGGNets和ResNets模型时是有效的。结果表明，VGGNets的能耗大幅降低了38.8%，ResNets降低了48.0%，这表明在脉冲神经网络中实现显著的效率提升具有潜力。这些发现有助于正在进行的提高脉冲神经网络性能的研究，促进其在资源受限环境中的部署。我们的代码可在GitHub上获取：https://github.com/hanebarla/BayesianSpikeFusion 。
