Fed-CASQ: Enhancing Class-Wise Accuracy in Pervasive Federated Learning with Class-Aware Scaling and Quantization

Dey, E.; Ravi, A.; Shinde, G.; Chugh, G.; Ghosh, I.; Misra, A.; Roy, N.

doi:10.1109/percom67906.2026.11524552

Publications › PerCom 2026

Fed-CASQ: Enhancing Class-Wise Accuracy in Pervasive Federated Learning with Class-Aware Scaling and Quantization

Dey, E., Ravi, A., Shinde, G., Chugh, G., Ghosh, I., Misra, A. & Roy, N.

PerCom 2026 Main Track Core A*ML / AISystems

DOI

Abstract

Federated Learning (FL) enables collaborative machine learning across decentralized devices and data sources, but resource constraints on pervasive devices necessitate efficient model compression. Existing approaches, such as quantization for on-device training, often degrade accuracy, especially for classes that are difficult to learn due to imbalance, poor-quality samples, or inherent complexity. This results in persistent accuracy gaps across classes. We propose Fed-CASQ's a novel framework that couples class-aware strategies into the quantization process to jointly improve efficiency and accuracy in pervasive FL. Unlike prior works that address quantization and imbalance separately, Fed-CASQ adaptively selects quantization levels based on device resources and leverages Layer-wise Relevance Propagation (LRP) to assess class-relevant convolutional neural network (CNN) filters on the client side. An adaptive weight scaling mechanism is then applied to amplify critical information for low-accuracy classes before aggregation. At the server, a complementary novel aggregation strategy mitigates global imbalance across clients, ensuring that underperforming classes receive proportional attention during model updates. We theoretically establish that Fed-CASQ achieves a convergence rate of ${\mathcal{O}}\left({\frac{{\kappa *\hat \sigma *\hat \delta }}{{\sqrt T }}}\right)$ under non-convex settings. We empirically establish that quantization directly influences the performance of under sampled (minority) classes. Experimental results further show that Fed-CASQ substantially narrows the performance gap for low-accuracy classes, improving their accuracy by ≈30%, while reducing training latency by over 56% on resource-constrained pervasive devices.

BibTeX

@inproceedings{dey2026fedcasq,
  title={Fed-CASQ: Enhancing Class-Wise Accuracy in Pervasive Federated Learning with Class-Aware Scaling and Quantization},
  author={Dey, E. and Ravi, A. and Shinde, G. and Chugh, G. and Ghosh, I. and Misra, A. and Roy, N.},
  booktitle={PerCom 2026},
  year={2026},
  url={https://doi.org/10.1109/percom67906.2026.11524552}
}