Title:
INFORMATION PROCESSING DEVICE AND INFORMATION PROCESSING METHOD
Document Type and Number:
WIPO Patent Application WO/2021/075107
Kind Code:
A1
Abstract:
An information processing device (100) is provided with: an acquisition unit (153) that acquires a machine learning model which has been subjected to reinforcement learning on the basis of a plurality of rewards weighted by the respective weights of the rewards such that the model outputs, upon input of first state information indicative of a first state, first behavior information indicative of a first behavior corresponding to the first state; a reception unit (151) that receives teaching data, which is a combination of second state information indicative of a second state and second behavior information indicative of a second behavior corresponding to the second state; and a display unit (156) that displays information pertaining to the reward weights estimated by training the machine learning model, in which the reward weights are used as connection coefficients of a part of the machine learning model, such that the model outputs the second behavior information included in the teaching data upon input of the second state information included in the teaching data and of values based on the reward weights.
Inventors:
KIMURA TOMOYA (JP)
Application Number:
PCT/JP2020/027416
Publication Date:
April 22, 2021
Filing Date:
July 14, 2020
Export Citation:
Assignee:
SONY CORP (JP)
International Classes:
G06N20/00
Domestic Patent References:
| WO2018110305A1 | 2018-06-21 |
Foreign References:
| JP2018181343A | 2018-11-15 |
Attorney, Agent or Firm:
SAKAI INTERNATIONAL PATENT OFFICE (JP)
Download PDF:
Previous Patent: METHOD AND DEVICE FOR FORMING LAYERED WORKPIECE
Next Patent: SIGNAL PROCESSING DEVICE AND METHOD, AND PROGRAM
Next Patent: SIGNAL PROCESSING DEVICE AND METHOD, AND PROGRAM