New Federated Learning Method Addresses Class Imbalance Across Distributed Data
Researchers have proposed FedBB, a federated learning framework that addresses class imbalance at three distinct levels—inter-case, inter-class, and inter-client—to improve model performance on non-IID data. The method combines a Positive Negative Balanced (PNB) loss function for local training with a Client Balanced Reweighting (CBR) mechanism during model aggregation. The work, accepted in Neurocomputing (2025), offers a potential baseline for both generic and personalized federated learning applications.
FedBB is a newly proposed federated learning (FL) algorithm designed to systematically resolve the non-IID (non-identically distributed) data problem by decomposing class imbalance into three levels: inter-case (imbalance within a single class), inter-class (imbalance between different classes), and inter-client (differences in data skewness across participating clients). To address these issues, the framework introduces two components: a Positive Negative Balanced (PNB) loss function that assigns higher weights to minority cases or classes during local training, supporting both multi-label and multi-class classification tasks, and a Client Balanced Reweighting (CBR) module that up-weights client models trained on less skewed datasets during global aggregation. Experiments conducted on X-ray and natural image datasets show FedBB outperforms existing algorithms in both accuracy and computational efficiency. Ablation studies confirm that PNB loss and CBR each independently contribute to overall performance gains. Notably, the method requires only limited statistical information from clients, which the authors argue is advantageous for privacy preservation. The paper has been accepted for publication in Neurocomputing (Volume 626, 2025) and is available as an author-accepted manuscript on arXiv.
What's missing
It is unclear how FedBB scales with very large numbers of clients or highly heterogeneous real-world deployments beyond the tested datasets. The degree to which the 'limited statistical information' requirement fully satisfies formal differential privacy guarantees is not explicitly analyzed.
What different sources said
- arXiv cs.LGCenter
Multi-Level Analyzation of Imbalance to Resolve Non-IID-Ness in Federated Learning
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.