“…Spiking neural networks (SNNs) [1] use unsupervised bio-inspired neurons and synaptic connections, trainable with either biological learning rules such as spike-timingdependent plasticity (STDP) [2] or supervised statistical learning algorithms such as surrogate gradient [3]. Empirical results on standard SNNs also show good performance for various tasks, including spatiotemporal data classification, [4,5], sequence-tosequence mapping [6], object detection [7,8], and universal function approximation [9,10]. An important motivation for the application of SNN in machine learning (ML) is the sparsity in the firing (activation) of the neurons, which reduces energy dissipation during inference [11].…”