US 12,755,178 B2 · Quantum federated learning system and method

Abstract

The present invention relates to a quantum federated learning system that performs federated learning on the basis of at least one observation value input from a single-hop offloading environment, and the system includes: a global server for initializing parameters of a quantum slimmable neural network (QSNN) model and transmitting the initialized quantum slimmable neural network model to at least one local device; and the at least one local device for inputting the at least one observation value into the initialized quantum slimmable neural network model to train the quantum slimmable neural network model, and transmitting the parameters of the trained quantum slimmable neural network model to the global server side. Through the system, the environmental epidemiology problems of the federated learning performed in conventional computing, such as communication channel conditions and energy limitations over time can be solved.

Claims10

Dependent claims

A quantum federated learning system that performs federated learning on the basis of at least one observation value input from a single-hop offloading environment, the system comprising:

a global server for initializing parameters of a quantum slimmable neural network (QSNN) model and transmitting the initialized quantum slimmable neural network model to at least one local device;
and
the at least one local device for inputting the at least one observation value into the initialized quantum slimmable neural network model to train the quantum slimmable neural network model,
wherein the at least one local device determines a channel state of the single-hop offloading environment based on a calculated reception throughput, and transmits at least one of an updated angle parameter and an updated axis parameter of the trained quantum slimmable neural network model to the global server side in a differential manner according to the determined channel state to minimize packet loss.

1 more element

A quantum federated learning method executed in a quantum federated learning system that performs federated learning, the method comprising the steps of:

initializing parameters of a quantum slimmable neural network (QSNN) model and transmitting the initialized quantum slimmable neural network model to at least one local device, by a global server;
receiving at least one observation value from a single-hop offloading environment, by the at least one local device;
inputting the at least one observation value into the initialized quantum slimmable neural network model to train the quantum slimmable neural network model, determining, by the at least one local device, a channel state of the single-hop offloading environment based on a calculated reception throughput;
and
transmitting the parameters of the trained quantum slimmable neural network model to the global server side, by the at least one local device,
wherein the transmitting includes transmitting at least one of an updated angle parameter and an updated axis parameter in a differential manner according to the determined channel state.

3 more elements

Dependencies are inferred from the claim text.

Quantum federated learning system and method

Life of this patent

Quantum federated learning system and method

Life of this patent

Abstract

Claims10