Reproduce “Exploiting Noise as a Resource for Computation and Learning in Spiking Neural Networks” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Gehua Ma, Rui Yan, Huajin Tang

The paper introduces noisy spiking neural networks (NSNNs) and noise-driven learning (NDL), using membrane-voltage randomness to make firing probabilistic and derive a learning signal from noise statistics. On DVS-CIFAR with STBP and ResNet-19, at spike-flip probability β = 0.04, accuracy was 58.99% ± 6.80 for the noisy model versus 40.32% ± 1.91 for the deterministic counterpart, supporting resilience under tested conditions.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.