New Training Method Enables Visible-Light Diffractive Neural Networks at Scale
Researchers developed a new training technique called differentiable beam-propagation (∂BPM) that overcomes a fundamental obstacle preventing diffractive neural networks from operating effectively in the visible light spectrum. The key insight is that the thin-layer approximation used in previous training methods fails not because of short wavelengths, but because visible-range materials require thick enough structures that internal light diffraction becomes significant. This advance could enable miniaturized, power-efficient optical processors for machine vision applications.
A team of researchers has identified and solved a critical problem limiting diffractive deep neural networks (D2NNs) in the visible spectrum. While D2NNs have shown promise in the terahertz regime as power-efficient optical processors, translating them to visible wavelengths—where most vision systems operate—has proven difficult. The researchers discovered that the standard thin-layer approximation used in training these networks breaks down at visible wavelengths not due to fabrication challenges, but because the low-refractive-index materials required (n ≈ 1.3-1.5) must be thick enough that internal diffraction effects become significant. Their solution, a differentiable beam-propagation layer (∂BPM), models each optical element as a finite-thickness volume and propagates light through it during training while maintaining fabrication compatibility. Testing on standard benchmarks (MNIST, Fashion-MNIST, CIFAR-100) showed substantial improvements, with full-wave validation raising classification accuracy from 50% to 90% without additional optimization.
What different sources said
- arXiv physicsCenter
Beyond the Thin-Layer Limit: Differentiable Volumetric Training for Visible-Range Diffractive 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.