Lernvorrichtung, Lernverfahren und Computerprogrammprodukt zum Trainieren

Details

Veröffentlichungs-Nr.

EP4092576

Aktenzeichen

EP22157720

Anmeldetag

21. Februar 2022

Veröffentlichung

22. Februar 2023

Rechtsraum

EP

IPC

G06N3/00G06N3/08G06N5/04G06N20/00

Abstract

A learning device 10A includes an acquisition unit 40A, a first action determination unit 40B, a second action determination unit 40C, and a selection unit 40D. The acquisition unit 40A acquires a current state of a mobile robot 20 as a device. The first action determination unit 40B learns a reinforcement learning model, and determines a first action of the mobile robot 20 on the basis of the current state and the reinforcement learning model. The second action determination unit 40C determines a second action of the mobile robot 20 on the basis of the current state and a first rule. The selection unit 40D selects one of the first action and the second action as a third action to be output to the mobile robot 20 according to a progress of learning of the reinforcement learning model.

Anmelder

Firma

KABUSHIKI KAISHA TOSHIBA

Land

🇯🇵 Japan

Fachgebiete