Learnable Channel Assignment Improves Forward-Only Convolutional Neural Networks
Researchers have proposed a learnable channel-class assignment mechanism for convolutional neural networks that operates without backpropagation, achieving state-of-the-art results among Forward-Forward (FF)-based models. The work builds on the biologically inspired Forward-Forward algorithm, which replaces gradient-based credit assignment with local, forward-only learning objectives. The findings suggest that adaptive channel specialization could make biologically plausible learning algorithms more competitive with standard deep learning training methods.
A new study posted to arXiv introduces a learnable channel-class assignment mechanism integrated into residual convolutional neural networks (CNNs) as an extension of the Forward-Forward (FF) algorithm, a biologically inspired alternative to backpropagation. Unlike prior FF-based CNN approaches that rely on static channel-class partitions, the proposed method allows channels to specialize adaptively based on data, guided by entropy and orthogonality regularization. The researchers also introduce a loss-aware layer contribution strategy that weights intermediate-layer predictions according to their validation performance during inference. Tested on CIFAR-10, CIFAR-100, and Tiny-ImageNet benchmarks, the method consistently outperforms existing forward-only approaches and sets a new state-of-the-art among FF-based models. The authors argue that these results substantially narrow the performance gap between forward-only learning and conventional backpropagation-trained networks, pointing toward more viable biologically plausible training paradigms for deep learning.
What's missing
The paper does not report wall-clock training time or computational cost comparisons relative to backpropagation, which is a key practical consideration for adoption. It is also unclear how the method scales to larger, more complex datasets beyond Tiny-ImageNet, or whether the learnable assignment mechanism introduces significant hyperparameter sensitivity. As a preprint, the work has not yet undergone peer review.
What different sources said
- arXiv cs.LGCenter
Forward-Only Convolutional Neural Networks with Learnable Channel-Class Assignment
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.