Title:
ニューラルネットワークアーキテクチャを探索する方法及び装置
Document Type and Number:
Japanese Patent JP7248190
Kind Code:
B2
Abstract:
A method and an apparatus for searching a neural network architecture comprising a backbone network and a feature network. The method comprises: a. forming a first search space for the backbone network and a second search space for the feature network; b. using a first controller to sample a backbone network model in the first search space, and using a second controller to sample a feature network model in the second search space; c. combining the first controller and the second controller by adding collected entropy and probability of the sampled backbone network model and feature network model to obtain a combined controller; d. using the combined controller to obtain a combined model; e. evaluating the combined model, and updating a combined model parameter according to an evaluation result; f. determining a verification accuracy of the updated combined model, and updating the combined controller according to the verification accuracy.
More Like This:
Inventors:
Jean Hoigan
Thou Yuan
Shun Son
Thou Yuan
Shun Son
Application Number:
JP2022500783A
Publication Date:
March 29, 2023
Filing Date:
July 15, 2019
Export Citation:
Assignee:
富士通株式会社
International Classes:
G06N3/0985; G06N3/0464; G06T7/00; G06V10/82
Domestic Patent References:
| JP2012243256A |
Foreign References:
| WO2019018375A1 |
Other References:
Golnaz Ghiasi et al.,"NAS-FPN: Learning Scalable Feature Pyramid Architecture for Object Detection",arXiv.org [online],arXiv:1904.07392v1,Cornell University,2019年04月,[令和5年2月3日検索], インターネット:
Barret Zoph et al.,"Learning Transferable Architectures for Scalable Image Recognition",arXiv.org [online],arXiv:1707.07012v4,Cornell University,2018年04月,[令和5年2月3日検索], インターネット:
Bichen Wu et al.,"FBNet: Hardware-Aware Efficient ConvNet Design via Differentiable Neural Architecture Search",arXiv.org [online],arXiv:1812.03443v3,Cornell University,2019年05月,[令和5年2月3日検索], インターネット:
Barret Zoph et al.,"Learning Transferable Architectures for Scalable Image Recognition",arXiv.org [online],arXiv:1707.07012v4,Cornell University,2018年04月,[令和5年2月3日検索], インターネット:
Bichen Wu et al.,"FBNet: Hardware-Aware Efficient ConvNet Design via Differentiable Neural Architecture Search",arXiv.org [online],arXiv:1812.03443v3,Cornell University,2019年05月,[令和5年2月3日検索], インターネット:
Attorney, Agent or Firm:
Tadashige Ito
Tadahiko Ito
Tadahiko Ito
Previous Patent: semiconductor circuit device
Next Patent: EQUIPMENT FOR HOT REPAIR OF FURNACE WALL OF KILN
Next Patent: EQUIPMENT FOR HOT REPAIR OF FURNACE WALL OF KILN