Inspired by detailed modelling of biological neurons, spiking neural networks (SNNs) are investigated as biologically plausible and high-performance models of neural computation. The sparse and binary communication between spiking neurons potentially enables powerful and energy-efficient neural networks. The performance of SNNs, however, has remained lacking compared with artificial neural networks. Here we demonstrate how an activity-regularizing surrogate gradient combined with recurrent networks of tunable and adaptive spiking neurons yields the state of the art for SNNs on challenging benchmarks in the time domain, such as speech and gesture recognition. This also exceeds the performance of standard classical recurrent neural networks and approaches that of the best modern artificial neural networks. As these SNNs exhibit sparse spiking, we show that they are theoretically one to three orders of magnitude more computationally efficient compared to recurrent neural networks with similar performance. Together, this positions SNNs as an attractive solution for AI hardware implementations.

受生物神经元详细建模的启发，脉冲神经网络 (SNN) 被研究为具有生物学合理性和高性能的神经计算模型。尖峰神经元之间的稀疏和二进制通信可能会实现强大且节能的神经网络。然而，与人工神经网络相比，SNN 的性能仍然有所欠缺。在这里，我们展示了活动正则化代理梯度如何与可调和自适应尖峰神经元的循环网络相结合，为 SNN 在时域具有挑战性的基准（例如语音和手势识别）上产生最先进的技术。这也超过了标准经典递归神经网络的性能，并接近了最好的现代人工神经网络的性能。由于这些 SNN 表现出稀疏的尖峰，我们表明，与具有相似性能的递归神经网络相比，它们在理论上的计算效率要高出一到三个数量级。总之，这将 SNN 定位为人工智能硬件实施的有吸引力的解决方案。