Abstract: The increasing demand for on-chip edge intelligence has motivated the exploration of algorithmic techniques and specialized hardware to reduce the computing energy of current machine learning models. In particular, deep spiking neural networks (SNNs) have gained interest because their event-driven hardware implementations can consume very low energy. However, minimizing average spiking activity and thus energy consumption while preserving accuracy in deep SNNs remains a significant challenge and opportunity. This paper proposes a novel two-step SNN compression technique to reduce their spiking activity while maintaining accuracy that involves compressing specifically-designed artificial neural networks (ANNs) that are then converted into the target SNNs. Our approach uses an ultra-high ANN compression technique that is guided by the attention-maps of an uncompressed meta-model. We then evaluate the firing threshold of each ANN layer and start with the trained ANN weights to perform a sparse-learning-based supervised SNN training to minimize the number of timesteps required while retaining compression. To evaluate the merits of the proposed approach, we performed experiments with variants of VGG and ResNet, on both CIFAR-10and CIFAR-100, and VGG16 on Tiny-ImageNet. SNN mod-els generated through the proposed technique yield state-of-the-art compression ratios of up to 33.4x with no significant drop in accuracy compared to baseline unpruned counterparts. As opposed to the existing SNN pruning methods we achieve up to 8.3x better compression with no drop inaccuracy. Moreover, compressed SNN models generated by our methods can have up to 12.2x better compute energy-efficiency compared to ANNs that have a similar number of parameters.
Authors: Souvik Kundu, Gourav Datta, Massoud Pedram, Peter A. Beerel (University of Southern California)