Title:
KNOWLEDGE TRACING DEVICE, METHOD, AND PROGRAM
Document Type and Number:
WIPO Patent Application WO/2021/064931
Kind Code:
A1
Abstract:
According to the present invention, a variational parameter determination unit 81 determines a variational parameter that specifies a position at which a likelihood function and the lower bound of a Gaussian-approximated likelihood function are in contact. A gradient-direction lower bound calculation unit 82 generates a likelihood function that is mono-dimensionalized in a gradient direction at the center of a prior distribution and calculates the lower bound of the generated likelihood function. A full-dimension lower bound calculation unit 83 sets the covariance in other than the gradient direction to a discretionary covariance and calculates the lower bound of the set covariance.
Inventors:
TAMANO HIROSHI (JP)
Application Number:
PCT/JP2019/039083
Publication Date:
April 08, 2021
Filing Date:
October 03, 2019
Export Citation:
Assignee:
NEC CORP (JP)
International Classes:
G06N7/00
Other References:
CHUN-KIT YEUNG; DIT-YAN YEUNG: "Addressing Two Problems in Deep Knowledge Tracing via Prediction-Consistent Regularization", ARXIV.ORG, vol. 1, no. 1806.02180v1, 6 June 2018 (2018-06-06), pages 1 - 10, XP080887737
WANG ZHIWEI, FENG XIAOQIN, TANG JILIANG, HUANG GALE YAN, LIU ZITAO: "Deep Knowledge Tracing with Side Information", ARXIV.ORG, vol. 11626, 1 September 2019 (2019-09-01), pages 303 - 308, XP055810893
WANG ZHIWEI, FENG XIAOQIN, TANG JILIANG, HUANG GALE YAN, LIU ZITAO: "Deep Knowledge Tracing with Side Information", ARXIV.ORG, vol. 11626, 1 September 2019 (2019-09-01), pages 303 - 308, XP055810893
Attorney, Agent or Firm:
IWAKABE Fuyuki et al. (JP)
Download PDF: