Learnable Leakage and Onset-Spiking Self-Attention in SNNs with Local Error Signals

Abstract

Spiking neural networks (SNNs) have garnered significant attention due to their computational patterns resembling biological neural networks. However, when it comes to deep SNNs, how to focus on critical information effectively and achieve a balanced feature transformation both temporally and spatially becomes a critical challenge. To address these challenges, our research is centered around two aspects: structure and strategy. Structurally, we optimize the leaky integrate-and-fire (LIF) neuron to enable the leakage coefficient to be learnable, thus making it better suited for contemporary applications. Furthermore, the self-attention mechanism is introduced at the initial time step to ensure improved focus and processing. Strategically, we propose a new normalization method anchored on the learnable leakage coefficient (LLC) and introduce a local loss signal strategy to enhance the SNN’s training efficiency and adaptability. The effectiveness and performance of our proposed methods are validated on the MNIST, FashionMNIST, and CIFAR-10 datasets. Experimental results show that our model presents a superior, high-accuracy performance in just eight time steps. In summary, our research provides fresh insights into the structure and strategy of SNNs, paving the way for their efficient and robust application in practical scenarios.

Benchmarking Artificial Neural Network Architectures for High-Performance Spiking Neural Networks

“…There has been a notable surge in research focused on SNNs in recent years [ 6 , 7 , 8 , 9 ]. The predominant catalyst for this increased attention is the energy-efficient operation characteristic of these networks, as highlighted in [ 10 ].…”