Title:
DATA CONVERSION LEARNING APPARATUS, PATTERN RECOGNITION DEVICE, DATA CONVERSION LEARNING METHOD, AND RECORDING MEDIUM
Document Type and Number:
WIPO Patent Application WO/2020/152774
Kind Code:
A1
Abstract:
This data conversion learning apparatus is provided with; a data conversion unit that receives, as input, source data and target data, and performs data conversion; a domain identification unit that identifies a domain from the converted data; a class identification unit that identifies a class from the converted data; a first deduction unit that deduces data of a class yet to appear on the basis of a domain certainty factor acquired from the domain identification unit; a second deduction unit that deduces data of a class yet to appear on the basis of a class certainty factor acquired from the class identification unit; a class identification learning unit that performs machine learning for class identification by using the data of the class yet to appear deduced by the first deduction unit and the source data and the target data which have been received as input; and a domain identification learning unit that performs machine learning for domain identification by using the data of the class yet to appear deduced by the second deduction unit, and the source data and the target data which have been received as input.
More Like This:
Inventors:
ISHII MASATO (JP)
TAKENOUCHI TAKASHI (JP)
SUGIYAMA MASASHI (JP)
TAKENOUCHI TAKASHI (JP)
SUGIYAMA MASASHI (JP)
Application Number:
PCT/JP2019/001842
Publication Date:
July 30, 2020
Filing Date:
January 22, 2019
Export Citation:
Assignee:
NEC CORP (JP)
International Classes:
G06N20/00
Other References:
YAROSLAV GANIN; EVGENIYA USTINOVA; HANA AJAKAN; PASCAL GERMAIN; HUGO LAROCHELLE; FRANÇOIS LAVIOLETTE; MARIO MARCHAND; VICTOR LEMPI: "Domain-adversarial training of neural networks", THE JOURNAL OF MACHINE LEARNING RESEARCH, vol. 17, 26 May 2016 (2016-05-26), pages 1 - 35, XP002789597
MASASHI SUGIYAMA, YAMADA MAKOTO, CHRISTOFFEL MARTHINUS, PLESSIS DU, LIU SONG: "Learning under non-stationarity: covariate shift adaptation, class-balance change adaptation and change detection", JOURNAL OF THE JAPAN STATISTICAL SOCIETY, JAPANESE ISSUE, vol. 44, no. 1, 26 September 2014 (2014-09-26), pages 113 - 136, XP055728459, ISSN: 2189-1478
SUGIYAMA MASASHI: "Big data analysis by density ration estimation", THE JOURNAL INSTITUTE OF ELECTRONICS INFORMATION AND COMMUNICATION ENGINEERS, vol. 97, no. 5, 1 May 2014 (2014-05-01), pages 353 - 358, ISSN: 0913-5693
MASASHI SUGIYAMA, YAMADA MAKOTO, CHRISTOFFEL MARTHINUS, PLESSIS DU, LIU SONG: "Learning under non-stationarity: covariate shift adaptation, class-balance change adaptation and change detection", JOURNAL OF THE JAPAN STATISTICAL SOCIETY, JAPANESE ISSUE, vol. 44, no. 1, 26 September 2014 (2014-09-26), pages 113 - 136, XP055728459, ISSN: 2189-1478
SUGIYAMA MASASHI: "Big data analysis by density ration estimation", THE JOURNAL INSTITUTE OF ELECTRONICS INFORMATION AND COMMUNICATION ENGINEERS, vol. 97, no. 5, 1 May 2014 (2014-05-01), pages 353 - 358, ISSN: 0913-5693
Attorney, Agent or Firm:
TANAI Sumio et al. (JP)
Download PDF: