Spiking neural networks (SNNs) are promising in a bio-plausible coding for spatio-temporal information and event-driven signal processing, which is very suited for energy-efficient implementation in neuromorphic hardware. However, the unique working mode of SNNs makes them more difficult to train than traditional networks. Currently, there are two main routes to explore the training of deep SNNs with high performance. The first is to convert a pre-trained ANN model to its SNN version, which usually requires a long coding window for convergence and cannot exploit the spatio-temporal features during training for solving temporal tasks. The other is to directly train SNNs in the spatio-temporal domain. But due to the binary spike activity of the firing function and the problem of gradient vanishing or explosion, current methods are restricted to shallow architectures and thereby difficult in harnessing large-scale datasets (e.g. ImageNet). To this end, we propose a threshold-dependent batch normalization (tdBN) method based on the emerging spatio-temporal backpropagation, termed "STBP-tdBN", enabling direct training of a very deep SNN and the efficient implementation of its inference on neuromorphic hardware. With the proposed method and elaborated shortcut connection, we significantly extend directly-trained SNNs from a shallow structure ( < 10 layer) to a very deep structure (50 layers). Furthermore, we theoretically analyze the effectiveness of our method based on "Block Dynamical Isometry" theory. Finally, we report superior accuracy results including 93.15 % on CIFAR-10, 67.8 % on DVS-CIFAR10, and 67.05% on ImageNet with very few timesteps. To our best knowledge, it's the first time to explore the directly-trained deep SNNs with high performance on ImageNet.
翻译:Spik 神经网络(SNNS)在用于时空空间信息和事件驱动信号处理的可生物化网络编码中很有希望,这种编码非常适合神经变形硬件的节能实施。然而,SNNS的独特工作模式使得它们比传统网络更难培训。目前,有两个主要途径可以探索深层 SNNS 高性能的培训。第一个途径是将经过预先训练的 ANN 模型转换为 SNN 版本,通常需要一个长的深度编码窗口,用于趋同,在培训中无法利用spatio-时空功能解决时间性任务。另一个途径是直接在神经变异域域内直接培训SNNNNNNS。目前的方法仅限于浅层结构,因此难以使用大型数据集(e.g.图像Net)。为此,我们建议基于新兴的Spotio-stal-stalalal 10-stalal-stal-deal-deformlation 10-deal-deal report S-S-S-ST-NBS-lal ral intal intal intal intal-deal intal-deal-deal intal intal intal intal intal intal-ILILILI.