US 12,423,618 B2
Federated learning system using synonym
Chih-Fan Hsu, Taipei (TW); Wei-Chao Chen, Taipei (TW); and Ming-Ching Chang, Taipei (TW)
Assigned to Inventec (Pudong) Technology Corporation, Shanghai (CN); and INVENTEC CORPORATION, Taipei (TW)
Filed by Inventec (Pudong) Technology Corporation, Shanghai (CN); and INVENTEC CORPORATION, Taipei (TW)
Filed on Mar. 17, 2022, as Appl. No. 17/697,104.
Claims priority of application No. 202210232854.7 (CN), filed on Mar. 9, 2022.
Prior Publication US 2023/0289653 A1, Sep. 14, 2023
Int. Cl. G06N 20/00 (2019.01)
CPC G06N 20/00 (2019.01) 10 Claims
OG exemplary drawing
 
1. A federated learning system using synonym, comprising:
a plurality of client devices, wherein each of the plurality of client devices comprises:
a first processor configured to execute an encoder, wherein the encoder removes a private portion of private data and encodes the private data after removing the private portion into a digest, and the first processor further trains a client model according to the private data after removing the private portion, the digest and a general model; and
a first communication circuit electrically connected to the first processor, and configured to transmit the digest and a client parameter of the client model, wherein the client parameter is associated with a weight of the client model; and
a moderator communicably connected to each of the plurality of client devices, wherein the moderator comprises:
a second communication circuit configured to send the general model to each of the plurality of client devices; and
a second processor electrically connected to the second communication circuit, and configured to determine an absent client device among the plurality of client devices and execute a synonym generator, wherein the synonym generator generates a synonym of the digest corresponding to the absent client device, the second processor further trains a replacement model according to the synonym and the digest corresponding to the absent client device, and performs an aggregation to generate an updated parameter to update the general model according to a replacement parameter of the replacement model and the client parameter of each of the plurality of client devices except the absent client device.