The patent badge is an abbreviated version of the USPTO patent document. The patent badge does contain a link to the full patent document.
The patent badge is an abbreviated version of the USPTO patent document. The patent badge covers the following: Patent number, Date patent was issued, Date patent was filed, Title of the patent, Applicant, Inventor, Assignee, Attorney firm, Primary examiner, Assistant examiner, CPCs, and Abstract. The patent badge does contain a link to the full patent document (in Adobe Acrobat format, aka pdf). To download or print any patent click here.
Patent No.:
Date of Patent:
Mar. 05, 2013
Filed:
Nov. 02, 2009
Yugo Ueda, Wako, JP;
Tadaaki Hasegawa, Wako, JP;
Soshi Iba, Wako, JP;
Koji Akatsuka, Wako, JP;
Norikazu Sugimoto, Kyoto, JP;
Yugo Ueda, Wako, JP;
Tadaaki Hasegawa, Wako, JP;
Soshi Iba, Wako, JP;
Koji Akatsuka, Wako, JP;
Norikazu Sugimoto, Kyoto, JP;
Honda Motor Co., Ltd., Tokyo, JP;
Advanced Telecommunications Research Institute International, Kyoto, JP;
Abstract
A reinforcement learning system () of the present invention utilizes a value of a first value gradient function (dV/dt) in the learning performed by a second learning device (), namely in evaluating a second reward (r(t)). The first value gradient function (dV/dt) is a temporal differential of a first value function (V) which is defined according to a first reward (r(t)) obtained from an environment and is served as a learning result given by a first learning device (). An action policy which should be taken by a robot (R) to execute a task is determined based on the second reward (r(t)).