Razor SNN: Efficient Spiking Neural Network with Temporal Embeddings

TL;DR

Razor SNN introduces a novel framework that prunes irrelevant event frames in spiking neural networks using temporal embeddings, significantly improving efficiency while maintaining competitive accuracy on multiple event-based benchmarks.

Contribution

The paper presents a new sparsification framework for SNNs that adaptively prunes event frames during inference using learned temporal embeddings, enhancing efficiency without sacrificing performance.

Findings

Achieves competitive accuracy on four event-based benchmarks.

Effectively reduces computational redundancy by pruning unnecessary event frames.

Demonstrates consistent efficiency improvements across diverse datasets.

Abstract

The event streams generated by dynamic vision sensors (DVS) are sparse and non-uniform in the spatial domain, while still dense and redundant in the temporal domain. Although spiking neural network (SNN), the event-driven neuromorphic model, has the potential to extract spatio-temporal features from the event streams, it is not effective and efficient. Based on the above, we propose an events sparsification spiking framework dubbed as Razor SNN, pruning pointless event frames progressively. Concretely, we extend the dynamic mechanism based on the global temporal embeddings, reconstruct the features, and emphasize the events effect adaptively at the training stage. During the inference stage, eliminate fruitless frames hierarchically according to a binary mask generated by the trained temporal embeddings. Comprehensive experiments demonstrate that our Razor SNN achieves competitive performance consistently on four events-based benchmarks: DVS 128 Gesture, N-Caltech 101, CIFAR10-DVS and SHD.