Struktur für ein Maschinenlernmodell zur Reservoirberechnung und Anwendung davon zur Vorhersage der Materialverschlechterung

Details

Veröffentlichungs-Nr.

EP4814803

Anmeldetag

6. Februar 2026

Veröffentlichung

30. September 2026

Rechtsraum

EP

IPC

G06N3/08, G06N3/044, G06N10/60

Abstract

In order to improve that analysis and predictions of complex tasks such as material degradation prediction the invention proposes a machine learning model that is suitable for reservoir computing and comprises a computer-implemented input module configured for receiving input data; a plurality of reservoir networks, each reservoir network configured for receving data from the input module and for outputting reservoir state data; and computer-implemented readout module that is configured for receiving the reservoir state data from the reservoirs and for outputting readout data based on the reservoir state data. The reservoir networks can be computer-implemented or physically-implemented and at least one of them may involve a plurality of quibits.

Anmelder

Firmen

Airbus (S.A.S.)
qBraid

Land

🇫🇷 Frankreich

Fachgebiete