Neuromorphic data, recording frameless spike events, have attracted considerable attention for the spatiotemporal information components and the event-driven processing fashion. Spiking neural networks (SNNs) represent a family of event-driven models with spatiotemporal dynamics for neuromorphic computing, which are widely benchmarked on neuromorphic data. Interestingly, researchers in the machine learning community can argue that recurrent (artificial) neural networks (RNNs) also have the capability to extract spatiotemporal features although they are not event-driven. Thus, the question of “what will happen if we benchmark these two kinds of models together on neuromorphic data” comes out but remains unclear.

In this work, we make a systematic study to compare SNNs and RNNs on neuromorphic data, taking the vision datasets as a case study. First, we identify the similarities and differences between SNNs and RNNs (including the vanilla RNNs and LSTM) from the modeling and learning perspectives. To improve comparability and fairness, we unify the supervised learning algorithm based on backpropagation through time (BPTT), the loss function exploiting the outputs at all timesteps, the network structure with stacked fully-connected or convolutional layers, and the hyper-parameters during training. Especially, given the mainstream loss function used in RNNs, we modify it inspired by the rate coding scheme to approach that of SNNs. Furthermore, we tune the temporal resolution of datasets to test model robustness and generalization. At last, a series of contrast experiments are conducted on two types of neuromorphic datasets: DVS-converted (N-MNIST) and DVS-captured (DVS Gesture). Extensive insights regarding recognition accuracy, feature extraction, temporal resolution and contrast, learning generalization, computational complexity and parameter volume are provided, which are beneficial for the model selection on different workloads and even for the invention of novel neural models in the future.

记录无框架尖峰事件的神经形态数据已受到时空信息组件和事件驱动的处理方式的极大关注。尖峰神经网络（SNN）代表了一系列事件驱动的模型，这些模型具有时态动力学，可用于神经形态计算，该模型广泛地基于神经形态数据进行基准测试。有趣的是，机器学习社区的研究人员可以辩称，尽管不是事件驱动的，但循环（人工）神经网络（RNN）也具有提取时空特征的能力。因此，出现了“如果将这两种模型一起作为神经形态数据进行基准测试会发生什么”的问题，但仍不清楚。

在这项工作中，我们以视觉数据集为例，对神经形态数据上的SNN和RNN进行了比较系统的研究。首先，我们从建模和学习的角度确定SNN和RNN（包括香草RNN和LSTM）之间的异同。为了提高可比性和公平性，我们统一了基于时间反向传播（BPTT）的监督学习算法，在所有时间步均利用输出的损失函数，具有堆叠的全连接层或卷积层的网络结构以及训练期间的超参数。特别是，考虑到RNN中使用的主流损失函数，我们在速率编码方案的启发下对其进行了修改，使其接近SNN。此外，我们调整数据集的时间分辨率以测试模型的鲁棒性和概括性。最后，在两种类型的神经形态数据集上进行了一系列对比实验：DVS转换（N-MNIST）和DVS捕获（DVS手势）。提供了有关识别精度，特征提取，时间分辨率和对比度，学习概括，计算复杂性和参数量的广泛见解，这对于在不同工作负载下进行模型选择，甚至对于将来发明新型神经模型都非常有用。