Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
PARAMETER ADJUSTING DEVICE, INFERENCE DEVICE, PARAMETER ADJUSTING METHOD, AND PARAMETER ADJUSTING PROGRAM
Document Type and Number:
WIPO Patent Application WO/2021/182000
Kind Code:
A1
Abstract:
A parameter adjusting device according to one aspect of the present invention computes the degree of correlation between a target inference task and each existing inference task in accordance with a similarity in purpose between the target inference task and each existing inference task, and, among a plurality of existing weights making up each existing weight set indicated by existing task information, determines a plurality of target weights making up a target weight set in accordance with the computed degree of correlation.

Inventors:
IMAI HIROSHI (JP)
YONETANI RYO (JP)
MIYAURA HIROYUKI (JP)
Application Number:
PCT/JP2021/005063
Publication Date:
September 16, 2021
Filing Date:
February 10, 2021
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
OMRON TATEISI ELECTRONICS CO (JP)
International Classes:
G06N20/00
Foreign References:
US20170200041A12017-07-13
JP2012026982A2012-02-09
Other References:
LI, Z. Q. ET AL.: "Heterogeneous defect prediction through multiple kernel learning and ensemble learning", PROCEEDINGS OF THE 2017 IEEE INTERNATIONAL CONFERENCE ON SOFTWARE MAINTENANCE AND EVOLUTION (ICSME, 22 September 2017 (2017-09-22), pages 91 - 102, XP033248468, ISBN: 978-1-5386-0992-7, DOI: 10.1109/ICSME.2017.19
LIN, C. K. ET AL.: "Exploring ensemble of models in taxonomy-based cross-domain sentiment classification", PROCEEDINGS OF THE 23RDACMINTERNATIONAL CONFERENCE ON INFORMATION AND KNOWLEDGE MANAGEMENT (CIKM'14, 7 November 2014 (2014-11-07), pages 1279 - 1288, XP058061118, ISBN: 978-1-4503-2598-1, DOI: 10.1145/2661829.2662071
DECENTRALIZED LEARNING TECHNIQUE ''DECENTRALIZED X, 11 March 2020 (2020-03-11), Retrieved from the Internet
See also references of EP 4120148A4
Attorney, Agent or Firm:
TACHIBANA, Kenji et al. (JP)
Download PDF: