Title:
OPTIMIZATION APPARATUS, OPTIMIZATION METHOD, AND PROGRAM FOR OPTIMIZATION
Document Type and Number:
Japanese Patent JP2023110297
Kind Code:
A
Abstract:
To allow even a small-scale edge computer to optimize a parameter by minimizing a loss function in a neural network, thereby saving memory and increasing a speed in the optimization processing.SOLUTION: In an optimizer OP that optimizes a parameter by minimizing a loss function corresponding to a prediction value and a ground truth value in a neural network, while limiting the number of quantization bits based on fixed-point numbers, logarithmic quantization is applied to a momentum mt-1 (mt is a momentum at a time t, and wt is a weight parameter), and a result obtained by multiplying a learning rate by a gradient of a loss function corresponding to the weight parameter is subtracted from a result of the logarithmic quantization, to output a result of the subtraction as a momentum mt, and, the momentum mt is added to a weight parameter wt-1, to be output as a weight parameter wt.SELECTED DRAWING: Figure 1
Inventors:
ASAI TETSUYA
YAMAGISHI YOSHIHARU
KANEKO TATSUYA
YAMAGISHI YOSHIHARU
KANEKO TATSUYA
Application Number:
JP2022011651A
Publication Date:
August 09, 2023
Filing Date:
January 28, 2022
Export Citation:
Assignee:
UNIV HOKKAIDO
International Classes:
G06N3/049; G06N3/08
Attorney, Agent or Firm:
Patent Attorney Corporation Intect International Patent Office
Mikawa Ltd.
Mikawa Ltd.
Previous Patent: Strength distribution color image display method for materials
Next Patent: Food and drink conveying equipment
Next Patent: Food and drink conveying equipment