Föderiertes Lernverfahren mit Datenschutzbewahrung, Zugehöriges Computerprogramm und Rahmen

Details

Veröffentlichungs-Nr.

EP4586149

Aktenzeichen

EP24305097

Anmeldetag

15. Januar 2024

Veröffentlichung

16. Juli 2025

Rechtsraum

EP

IPC

G06N3/098G06N3/096G06N3/0464// G06N3/084G06N5/045

Abstract

The invention concerns a method (20) for performing privacy-preserving federated learning, the method comprising the steps: - training (22) an object detection model based on a central training dataset (12) to obtain a preliminary model (30); - for each of N local nodes (8), training (24) a respective copy of the preliminary model based on a respective private training dataset (14), thereby obtaining a local model (32); - computing (26) an average model (34) as an average of N intermediate models (38), each depending on a respective local model (32); and - generating (28) a central model (36) based: • on the preliminary model (30) and on an average output of the average model (34) based on a predetermined public dataset (10) as input; or • on the average model (34) and a local output of each local model (32) based on the public dataset (10) as input.

Anmelder

Firma

BULL SAS

Land

🇫🇷 Frankreich

Fachgebiete