English

Global Layers: Non-IID Tabular Federated Learning

Machine Learning 2023-06-01 v1

Abstract

Data heterogeneity between clients remains a key challenge in Federated Learning (FL), particularly in the case of tabular data. This work presents Global Layers (GL), a novel partial model personalization method robust in the presence of joint distribution P(X,Y)P(X,Y) shift and mixed input/output spaces X×YX \times Y across clients. To the best of our knowledge, GL is the first method capable of supporting both client-exclusive features and classes. We introduce two new benchmark experiments for tabular FL naturally partitioned from existing real world datasets: i) UCI Covertype split into 4 clients by "wilderness area" feature, and ii) UCI Heart Disease, SAHeart, UCI Heart Failure, each as clients. Empirical results in these experiments in the full-participant setting show that GL achieves better outcomes than Federated Averaging (FedAvg) and local-only training, with some clients even performing better than their centralized baseline.

Keywords

Cite

@article{arxiv.2305.19290,
  title  = {Global Layers: Non-IID Tabular Federated Learning},
  author = {Yazan Obeidi},
  journal= {arXiv preprint arXiv:2305.19290},
  year   = {2023}
}

Comments

Pre-print, under review. 24 pages, 17 tables, 3 figures. For experiment code see: https://github.com/transferFL/gl/

R2 v1 2026-06-28T10:51:03.658Z