Federated Learning (FL) is plagued by two key challenges: high communication overhead and performance collapse on heterogeneous (non-IID) data. Analytic FL (AFL) provides a single-round, data distribution invariant solution, but is limited to linear models. Subsequent non-linear approaches, like DeepAFL, regain accuracy but sacrifice the single-round benefit. In this work, we break this trade-off. We propose SAFLe, a framework that achieves scalable non-linear expressivity by introducing a structured head of bucketed features and sparse, grouped embeddings. We prove this non-linear architecture is mathematically equivalent to a high-dimensional linear regression. This key equivalence allows SAFLe to be solved with AFL's single-shot, invariant aggregation law. Empirically, SAFLe establishes a new state-of-the-art for analytic FL, significantly outperforming both linear AFL and multi-round DeepAFL in accuracy across all benchmarks, demonstrating a highly efficient and scalable solution for federated vision.
@article{arxiv.2512.03336,
title = {Single-Round Scalable Analytic Federated Learning},
author = {Alan T. L. Bacellar and Mustafa Munir and Felipe M. G. França and Priscila M. V. Lima and Radu Marculescu and Lizy K. John},
journal= {arXiv preprint arXiv:2512.03336},
year = {2026}
}
Comments
To appear in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026