FedHENet: A Frugal Federated Learning Framework for Heterogeneous Environments

Alejandro Dopico-Castro, Oscar Fontenla-Romero, Bertha Guijarro-Berdiñas, Amparo Alonso-Betanzos, Iván Pérez Digón
2/13/2026
cs.CVcs.LG

Official code: AlejandroDopico2/FedHENet (3 stars, MIT, PyTorch, last commit 2026-02-16).

Abstract

Federated Learning (FL) enables collaborative training without centralizing data, essential for privacy compliance in real-world scenarios involving sensitive visual information. Most FL approaches rely on expensive, iterative deep network optimization, which still risks privacy via shared gradients. In this work, we propose FedHENet, extending the FedHEONN framework to image classification. By using a fixed, pre-trained feature extractor and learning only a single output layer, we avoid costly local fine-tuning. This layer is learned by analytically aggregating client knowledge in a single round of communication using homomorphic encryption (HE). Experiments show that FedHENet achieves competitive accuracy compared to iterative FL baselines while demonstrating superior stability performance and up to 70\% better energy efficiency. Crucially, our method is hyperparameter-free, removing the carbon footprint associated with hyperparameter tuning in standard FL. Code available in https://github.com/AlejandroDopico2/FedHENet/

View on arXivView PDF

Code Implementations(1)

AlejandroDopico2/FedHENetOfficial100%
PyTorchMIT

🧠 Official implementation of FedHENet: A frugal, one-shot federated learning framework for heterogeneous environments. Privacy-preserving via HE, energy-efficient, and hyperparameter-free. Accepted at ESANN 2026.

30PythonSep 24, 20258 months agoMIT
federated-learningfrugal-aigreen-aihomomorphic-encryptionmachine-learning+3 more

Similar Papers

Benchmarking FedAvg and FedCurv for Image Classification Tasks

87% match1 repo52023

Optimizing Federated Learning by Entropy-Based Client Selection

85% match2024

FLUE: Federated Learning with Un-Encrypted model weights

85% match2024

FedML-HE: An Efficient Homomorphic-Encryption-Based Privacy-Preserving Federated Learning System

84% match1 repo4.0k2023

Scalable Federated Learning for Clients with Different Input Image Sizes and Numbers of Output Categories

84% match2023

Federated Learning with Workload Reduction through Partial Training of Client Models and Entropy-Based Data Selection

84% match2024

Cite this paper

@article{dopicocastro2026fedhenet,
  title  = {FedHENet: A Frugal Federated Learning Framework for Heterogeneous Environments},
  author = {Alejandro Dopico-Castro and Oscar Fontenla-Romero and Bertha Guijarro-Berdiñas and Amparo Alonso-Betanzos and Iván Pérez Digón},
  year   = {2026},
  eprint = {2602.13024},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2602.13024v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2602.13024.svg)](https://codeofpaper.com/paper/2602.13024)

Discussion