New Training Method Improves Deep Spiking Neural Networks with Adaptive Asymmetric Surrogate Gradients
Researchers have proposed A2SG, a unified framework using adaptive and asymmetric surrogate gradients to address key training challenges in deep spiking neural networks (SNNs). SNNs are biologically inspired neural networks that struggle to train effectively due to sharp loss landscapes and temporal inconsistency arising from non-differentiable spike functions. The work offers a principled method to improve both accuracy and energy efficiency across multiple SNN architectures and tasks.
A team of researchers has introduced A2SG (Adaptive and Asymmetric Surrogate Gradients), a framework designed to overcome two core difficulties in training deep spiking neural networks: sharp loss landscapes and temporal gradient inconsistency. The adaptive component adjusts an effective gradient window for spatio-temporal adaptation, reducing spatial gradient variation and preserving directional consistency across time steps. The asymmetric component assigns larger gradients to neurons with higher membrane potentials, reflecting biological neuronal dynamics, and the authors mathematically prove this yields lower gradient variation than symmetric alternatives. The paper further establishes a theoretical link between local gradient variation and loss landscape curvature, explaining why A2SG tends to converge to flatter minima and generalize better. Experiments were conducted on CNN-based and Transformer-based SNNs across image classification on both static and neuromorphic datasets, as well as segmentation tasks, with consistent accuracy and energy efficiency improvements reported. The work was accepted at ICML 2026 and code has been made publicly available.
What's missing
The paper does not report direct comparisons of absolute energy consumption figures against conventional artificial neural networks (ANNs), which would contextualize the practical energy efficiency gains.
What different sources said
- arXiv cs.LGCenter
A2SG:Adaptive and Asymmetric Surrogate Gradients for Training Deep Spiking Neural Networks
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.