Title:
INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, AND INFORMATION PROCESSING PROGRAM
Document Type and Number:
WIPO Patent Application WO/2022/195691
Kind Code:
A1
Abstract:
In the present invention, a storage unit (11) stores: a plurality of labeled data in which target data and labels representing correct responses are associated with one another; a plurality of unlabeled data which are target data not associated with the correct responses; and a deep learning model (110). A pseudo label generation unit (12) generates a pseudo label on the basis of the unlabeled data and the deep learning model. On the basis of the pseudo label and the labels included in the labeled data, a loss calculation unit (14) uses the deep learning model (110) to calculate loss for when identification of the unlabeled data is performed and for when identification of the labeled data is performed. An update unit (15) updates the deep learning model (110) on the basis of the loss calculated by the loss calculation unit (14).
Inventors:
OBINATA YUYA (JP)
YAMAMOTO TAKUMA (JP)
YAMAMOTO TAKUMA (JP)
Application Number:
PCT/JP2021/010452
Publication Date:
September 22, 2022
Filing Date:
March 15, 2021
Export Citation:
Assignee:
FUJITSU LTD (JP)
International Classes:
G06N3/08; G06N20/00
Other References:
DONG-HYUN LEE: "Pseudo-Label : The Simple and Efficient Semi-Supervised Learning Method for Deep Neural Networks", ICML 2013 WORKSHOP : CHALLENGES IN REPRESENTATION LEARNING (WREPL), ATLANTA, GEORGIA, USA, 16 June 2013 (2013-06-16), Atlanta, Georgia, USA , XP055716966, Retrieved from the Internet [retrieved on 20200721]
XIAOMENG XIN, JINJUN WANG, RUJXIE, SANPINZHOU, WENLI HUANG, NANNING ZHENG: "Semi-supervised person re- identification using multi-view clustering", PATTERN RECOGNITION, vol. 88, 19 November 2019 (2019-11-19), pages 285 - 297, XP055972201
SAITO KUNIAKI, USHIKU YOSHITAKA, HARADA TATSUYA: "Asymmetric Tri-training for Unsupervised Domain Adaptation", ARXIV.ORG, 13 May 2017 (2017-05-13), pages 1 - 12, XP055972209, Retrieved from the Internet [retrieved on 20221018]
YUKI M. ASANOCHRISTIAN RUPPRECHTANDREA VEDALDI: "Self-labelling via simultaneous clustering and representation learning", ICLR2020, 20 August 2020 (2020-08-20)
See also references of EP 4310734A4
XIAOMENG XIN, JINJUN WANG, RUJXIE, SANPINZHOU, WENLI HUANG, NANNING ZHENG: "Semi-supervised person re- identification using multi-view clustering", PATTERN RECOGNITION, vol. 88, 19 November 2019 (2019-11-19), pages 285 - 297, XP055972201
SAITO KUNIAKI, USHIKU YOSHITAKA, HARADA TATSUYA: "Asymmetric Tri-training for Unsupervised Domain Adaptation", ARXIV.ORG, 13 May 2017 (2017-05-13), pages 1 - 12, XP055972209, Retrieved from the Internet
YUKI M. ASANOCHRISTIAN RUPPRECHTANDREA VEDALDI: "Self-labelling via simultaneous clustering and representation learning", ICLR2020, 20 August 2020 (2020-08-20)
See also references of EP 4310734A4
Attorney, Agent or Firm:
SAKAI INTERNATIONAL PATENT OFFICE (JP)
Download PDF:
Previous Patent: LOG DATA ANALYSIS DEVICE, LOG DATA ANALYSIS METHOD, AND PROGRAM
Next Patent: DIGITAL-TO-ANALOG CONVERTER
Next Patent: DIGITAL-TO-ANALOG CONVERTER