Workload-Balanced Pruning for Sparse Spiking Neural Networks

TL;DR

This paper introduces u-Ticket, a workload-aware pruning method for sparse SNNs that ensures optimal hardware utilization, significantly reducing latency and energy consumption on resource-constrained devices.

Contribution

The paper proposes u-Ticket, a novel pruning approach that monitors and adjusts weights during LTH-based pruning to achieve perfect hardware utilization in sparse SNNs.

Findings

u-Ticket guarantees up to 100% hardware utilization.

Reduces latency by up to 76.9%.

Decreases energy cost by up to 63.8%.

Abstract

Pruning for Spiking Neural Networks (SNNs) has emerged as a fundamental methodology for deploying deep SNNs on resource-constrained edge devices. Though the existing pruning methods can provide extremely high weight sparsity for deep SNNs, the high weight sparsity brings a workload imbalance problem. Specifically, the workload imbalance happens when a different number of non-zero weights are assigned to hardware units running in parallel. This results in low hardware utilization and thus imposes longer latency and higher energy costs. In preliminary experiments, we show that sparse SNNs (~98% weight sparsity) can suffer as low as ~59% utilization. To alleviate the workload imbalance problem, we propose u-Ticket, where we monitor and adjust the weight connections of the SNN during Lottery Ticket Hypothesis (LTH) based pruning, thus guaranteeing the final ticket gets optimal utilization when deployed onto the hardware. Experiments indicate that our u-Ticket can guarantee up to 100% hardware utilization, thus reducing up to 76.9% latency and 63.8% energy cost compared to the non-utilization-aware LTH method.