Title:
LEARNING SYSTEM, LEARNING METHOD, AND COMPUTER-READABLE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2024/034077
Kind Code:
A1
Abstract:
Provided are a learning system, a learning method, and a program that prevent the training data used for federated learning from being inferred. A learning system (10) comprises an acquisition unit (11) that acquires obfuscation data for obfuscating prescribed data included in data sets owned by respective organizations, a training unit (12) that trains local models by machine learning that uses the obfuscation data and the data sets, and a generation unit (13) that generates a federated learning global model on the basis of the local models.
Inventors:
AKAGAWA TAKESHI (JP)
Application Number:
PCT/JP2022/030623
Publication Date:
February 15, 2024
Filing Date:
August 10, 2022
Export Citation:
Assignee:
NEC CORP (JP)
International Classes:
G06N3/08
Domestic Patent References:
| WO2015155896A1 | 2015-10-15 |
Foreign References:
| JP2021056677A | 2021-04-08 | |||
| JP2020052689A | 2020-04-02 | |||
| US20200005133A1 | 2020-01-02 |
Other References:
NAKAI TSUNATO, DAISUKE SUZUKI, TAKESHI FUJINO: "Towards Trained Model Protection for Byzantine-Robust Federated Learning", 2022 CRYPTOGRAPHY AND INFORMATION SECURITY SYMPOSIUM PROCEEDINGS, 1 January 2022 (2022-01-01), XP093139206
Attorney, Agent or Firm:
IEIRI Takeshi (JP)
Download PDF:
Previous Patent: SUPERABRASIVE GRAIN AND GRINDSTONE
Next Patent: DISPLAY CONTROL METHOD AND DISPLAY CONTROL DEVICE
Next Patent: DISPLAY CONTROL METHOD AND DISPLAY CONTROL DEVICE