QB-LIF: Learnable-Scale Quantized Burst Neurons for Efficient SNNs

TL;DR

The paper introduces QB-LIF, a learnable-scale quantized burst neuron model for spiking neural networks that enhances information throughput and accuracy in deep, low-latency architectures while maintaining hardware efficiency.

Contribution

It proposes a novel QB-LIF neuron with trainable quantization scale and an absorbable scale strategy, enabling adaptive spiking resolution and efficient inference in SNNs.

Findings

Outperforms binary and fixed-burst SNNs on multiple benchmarks.

Achieves higher accuracy with ultra-low latency.

Maintains neuromorphic compatibility during inference.

Abstract

Binary spike coding enables sparse and event-driven computation in spiking neural networks (SNNs), yet its 1-bit-per-timestep representation fundamentally limits information throughput. This bottleneck becomes increasingly restrictive in deep architectures under short simulation horizons. We propose the Quantized Burst-LIF (QB-LIF) neuron, which reformulates burst spiking as a saturated uniform quantization of membrane potentials with a learnable scale. Instead of relying on predefined multi-threshold structures, QB-LIF treats the quantization scale as a trainable parameter, allowing each layer to autonomously adapt its spiking resolution to the underlying membrane-potential statistics. To preserve hardware efficiency, we introduce an absorbable scale strategy that folds the learned quantized scale into synaptic weights during inference, maintaining a strict accumulate-only (AC) execution paradigm. To enable stable optimization in the discrete multi-level space, we further design ReLSG-ET, a rectified-linear surrogate gradient with exponential tails that sustains gradient flow across burst intervals. Extensive experiments on static (CIFAR-10/100, ImageNet) and event-driven (CIFAR10-DVS, DVS128-Gesture) benchmarks demonstrate that QB-LIF consistently outperforms binary and fixed-burst SNNs, achieving higher accuracy under ultra-low latency while preserving neuromorphic compatibility.