B:\PAPER.TXT

SQAMIA 2026

B:\> type PAPER.TXT

PUBLISHEDSQAMIA 2026CEUR-WS Vol-4336 · 2026-11-19

Secure and Reversible Federated Learning for IoT-Driven Food Supply Chain Monitoring

Mihailo Ilić,  Nemanja Gabor,  Mirjana Ivanović,  Milenko Tošić

VENUE
13th Workshop on Software Quality Analysis, Monitoring, Improvement, and Applications
WHERE
Niš, Serbia · Sep 16–18, 2026
IN
CEUR Workshop Proceedings, Vol-4336 · published 2026-11-19
LICENSE
CC BY 4.0

B:\> head abstract

Food quality is checked at the edge with AISpectra units that read ultraviolet, visible and near-infrared spectra, on our own extra virgin olive oil adulteration dataset. Models train with federated learning, so supply-chain actors never share raw data. When a participant withdraws, or turns out to be poisoning the model, Fast-FedUL federated unlearning removes its contribution without retraining from scratch.

keywords: federated learning · supply chains · federated unlearning · IoT

B:\WORK.LOG2 AT HEAD 2026.06 INVT ML Engineer2025.12 VizLore R&D2025.09 Synechron2022.08 VegaITOPEN A:\EDU.SYSMSC · BSC 2024 MSc Computer Science2021–2024 BSc Computer ScienceFaculty of Sciences, Novi Sad+ 9 coursesOPEN A:\SKILLS.ENVENV --WHERE LANG= Python C Java TypeScriptML= PyTorch Flower LeRobotGPU= WGPU/WGSL MPIWEB= DATA= OPS= ...OPEN