Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
SYSTEM FOR GENERATING FEATURE VECTORS
Document Type and Number:
WIPO Patent Application WO/2020/218314
Kind Code:
A1
Abstract:
One or more processors use a machine learning model to generate a feature vector of an anchor sample, a feature vector of a positive sample, and a feature vector of a negative sample, and trains a machine learning model so as to satisfy a predefined condition. The condition is such that the distance between the feature vector of the anchor sample and the feature vector of the positive sample is less than the distance between the feature vector of the anchor sample and the feature vector of the negative sample, and a range that should be satisfied by the distance between the feature vector of the anchor sample and the feature vector of the negative sample is defined on the basis of a predefined semantic distance between a first class and a second class in a semantic space.

Inventors:
KLINKIGT MARTIN (JP)
CHHABRA MOHIT (JP)
HIROIKE ATSUSHI (JP)
MURAKAMI TOMOKAZU (JP)
Application Number:
PCT/JP2020/017270
Publication Date:
October 29, 2020
Filing Date:
April 21, 2020
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
HITACHI LTD (JP)
International Classes:
G06F16/583; G06F16/908
Foreign References:
US20170228641A12017-08-10
US20190065957A12019-02-28
Other References:
NI, JIAZHI ET AL.: "Fine-grained Patient Similarity Measuring using Deep Metric Learning, Proceedings of the 2017 ACM on Conference on Information and Knowledge Management (CIKM), ACM (Association for Computing Machinery", ACM DIGITA LIBRARY, 6 November 2017 (2017-11-06), pages 1189 - 1198, XP055758902, Retrieved from the Internet [retrieved on 20200622]
Attorney, Agent or Firm:
TOU-OU PATENT FIRM (JP)
Download PDF: