Title:
LEARNING DEVICE, LEARNING METHOD, AND RECORDING MEDIUM
Document Type and Number:
WIPO Patent Application WO/2023/047562
Kind Code:
A1
Abstract:
In this learning device, an estimation means performs inference for training data using an inference model, and outputs a class score. A weight calculation means calculates a weight from the output class score by using a weight function that increases more rapidly than a linear increase when the class score is too large or too small. A weight sum calculation means calculates a sum of weights for a mini-batch including a predetermined number of pieces of training data. A regularization term calculation means calculates a regularization term by applying, to the sum, a rescaling function defined to be a monotonously increasing function that increases more slowly than a linear increase. An optimization means optimizes the inference model by using a total loss including the regularization term.
Inventors:
YOSHIDA SHUHEI (JP)
Application Number:
PCT/JP2021/035277
Publication Date:
March 30, 2023
Filing Date:
September 27, 2021
Export Citation:
Assignee:
NEC CORP (JP)
International Classes:
G06N20/00
Domestic Patent References:
| WO2021144943A1 | 2021-07-22 |
Foreign References:
| JP2021502626A | 2021-01-28 |
Other References:
CLARA MEISTER; ELIZABETH SALESKY; RYAN COTTERELL: "Generalized Entropy Regularization or: There's Nothing Special about Label Smoothing", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 2 May 2020 (2020-05-02), 201 Olin Library Cornell University Ithaca, NY 14853 , XP081657295
Attorney, Agent or Firm:
NAKAMURA, Toshinobu et al. (JP)
Download PDF:
Previous Patent: ELEVATOR DEVICE
Next Patent: INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, AND PROGRAM
Next Patent: INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, AND PROGRAM