Title:
MODEL GENERATION SYSTEM, MODEL GENERATION METHOD, AND MODEL GENERATION PROGRAM
Document Type and Number:
WIPO Patent Application WO/2018/150445
Kind Code:
A1
Abstract:
A kernel function creation unit 81 defines a first kernel function using a two-dimensional feature representation representing a combination of two data features. A model learning unit 82 defines a linear model including the inner product of a map used by the first kernel function and a first weight, and performs learning using the defined linear model. A component expansion unit 83 expands a learned linear model and thereby defines an expanded component representation, which is a new data component representation. An expanded model generation unit 84 generates an expanded model including the inner product of expanded component representation data and a second weight.
Inventors:
NAKADAI SHINJI (JP)
ZHANG HAO (JP)
ZHANG HAO (JP)
Application Number:
PCT/JP2017/005248
Publication Date:
August 23, 2018
Filing Date:
February 14, 2017
Export Citation:
Assignee:
NEC CORP (JP)
International Classes:
G06N99/00
Foreign References:
| JP2010086466A | 2010-04-15 | |||
| US20150032682A1 | 2015-01-29 |
Other References:
CAO, HUI ET AL.: "Approximate RBF Kernel SVM and Its Applications in Pedestrian Classification", THE 1ST INTERNATIONAL WORKSHOP ON MACHINE LEARNING FOR VISION-BASED MOTION ANALYSIS, October 2008 (2008-10-01), XP055537619, Retrieved from the Internet [retrieved on 20170316]
Attorney, Agent or Firm:
IWAKABE Fuyuki et al. (JP)
Download PDF: