Abstract
Spiking neural networks (SNNs), a variant of artificial neural networks (ANNs) with the benefit of energy efficiency, have achieved the accuracy close to its ANN counterparts, on benchmark datasets such as CIFAR10/100 and ImageNet. However, comparing with frame-based input (e.g., images), event-based inputs from e.g., Dynamic Vision Sensor (DVS) can make a better use of SNNs thanks to the SNNs' asynchronous working mechanism. In this paper, we strengthen the marriage between SNNs and event-based inputs with a proposal to consider anytime optimal inference SNNs, or AOI-SNNs, which can terminate anytime during the inference to achieve optimal inference result. Two novel optimisation techniques are presented to achieve AOI-SNNs: a regularisation and a cutoff. The regularisation enables the training and construction of SNNs with optimised performance, and the cutoff technique optimises the inference of SNNs on event-driven inputs. We conduct an extensive set of experiments on multiple benchmark event-based datasets, including CIFAR10-DVS, N-Caltech101 and DVS128 Gesture. The experimental results demonstrate that our techniques are superior to the state-of-the-art with respect to the accuracy and latency.
Abstract (translated)
喷射神经网络(SNNs)是一种以能源效率为优势的人工神经网络(ANNs)的变体,已经在诸如CIFAR10/100和ImageNet等基准数据集上实现了与ANN相媲美的精度。然而,与帧输入(例如图像)相比,来自动态视觉传感器(DVS)的事件输入可以通过SNNs的异步工作机制更好地利用SNNs。在本文中,我们提出了一种建议,考虑考虑 anytime optimal inference SNNs,或AOI-SNNs,能够在 inference 期间任意时间终止以获得最优Inference结果。为了实现AOI-SNNs,我们介绍了两个新的优化技术: regularization 和 cutoff。 regularization 使得训练和构建具有最优性能的SNNs 变得容易, cutoff 技术优化了基于事件输入的SNNs的Inference。我们在多个基准事件输入数据集上进行了广泛的实验,包括CIFAR10-DVS、N-Caltech101和DVS128Gesture数据集。实验结果表明,我们的技术在精度和延迟方面比当前技术水平更高。
URL
https://arxiv.org/abs/2301.09522
