Federal learning-based model training method, device and equipment and storage medium

Abstract

The invention discloses a federated learning-based model training method, device and equipment and a storage medium. According to the invention, a broadcast training request is sent to each client device through a server, and each client device determines the bid of the respective client device according to the received broadcast training request and a bidding strategy, and sends the bid of the respective client device to the server; then the server selects a preset number of target client devices according to the received bid of each client device, and sends identifiers of the target client devices and a first global model to each client device; if the identifier of the client device is matched with the identifier of the target client device, the client device serves as the target client device, and the target client device trains a first global model according to a local sample data set of the target client device; the problem of low model training efficiency based on federal learning in the prior art is solved.

Description

technical field

[0001] The present invention relates to the field of model training based on federated learning, in particular to a model training method based on federated learning, a model training device based on federated learning, a model training device based on federated learning, and a computer-readable storage medium. Background technique

[0002] In the era of big data, the problem of data islands has become increasingly prominent. Federated learning, as a technology that tries to break data islands, has attracted widespread attention once it was proposed; among them, federated learning is a special distributed machine learning solution, that is, the central server cooperates with multiple parties to jointly train machine learning models. At present, in the federated learning scheme, when selecting devices to participate in training, most of them adopt a random strategy; however, in the system environment with heterogeneous data, the resource allocation of client d...

Examples

Embodiment Construction

[0065] It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

[0066] See figure 1 as shown, figure 1 It is a schematic structural diagram of a model training device based on federated learning in the hardware operating environment involved in the solution of the embodiment of the present invention.

[0067] It should be clear that, figure 1 The federated learning-based model training device shown in may be a federated learning-based model training device applied to a server or a client.

[0068] Wherein, when the model training device based on federated learning is a model training device based on federated learning applied to the server, the model training device based on federated learning includes: at least one processor 101, a memory 102, and a A model training program based on federated learning running on the server, and the model training program based on federated lea...

technical field

[0001] The present invention relates to the field of model training based on federated learning, in particular to a model training method based on federated learning, a model training device based on federated learning, a model training device based on federated learning, and a computer-readable storage medium. Background technique

[0002] In the era of big data, the problem of data islands has become increasingly prominent. Federated learning, as a technology that tries to break data islands, has attracted widespread attention once it was proposed; among them, federated learning is a special distributed machine learning solution, that is, the central server cooperates with multiple parties to jointly train machine learning models. At present, in the federated learning scheme, when selecting devices to participate in training, most of them adopt a random strategy; however, in the system environment with heterogeneous data, the resource allocation of client d...

Examples

Embodiment Construction

[0065] It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

[0066] See figure 1 as shown, figure 1 It is a schematic structural diagram of a model training device based on federated learning in the hardware operating environment involved in the solution of the embodiment of the present invention.

[0067] It should be clear that, figure 1 The federated learning-based model training device shown in may be a federated learning-based model training device applied to a server or a client.

[0068] Wherein, when the model training device based on federated learning is a model training device based on federated learning applied to the server, the model training device based on federated learning includes: at least one processor 101, a memory 102, and a A model training program based on federated learning running on the server, and the model training program based on federated lea...

Images