Systems and methods for federated model validation and data verification

Patent metadata

Field	Value
Publication number	US-12386979-B2
Application number	US-202318156607-A
Country	US
Kind code	B2
Filing date	Jan 19, 2023
Priority date	Jan 20, 2022
Publication date	Aug 12, 2025
Grant date	Aug 12, 2025

Abstract

Systems and methods for federated model validation and data verification are disclosed. A method may include: (1) receiving, by a local computer program executed by client system, a federated machine learning model from a federated model server; (2) testing, by the local computer program and using a policy service, the federated machine learning model for vulnerabilities to attacks; (3) accepting, by the local computer program, the federated machine learning model in response to the federated machine learning model passing the testing; (4) training, by the local computer program, the federated machine learning model using input data comprising local data and outputting training parameters; (5) identifying, by the local computer program using the policy service, accidental leakage and/or contamination by comparing the training parameters to the input data; and (6) providing, by the local computer program, the training parameters to the federated model server.

First claim

What is claimed is:

1. A method for federated model validation and data verification, comprising:
receiving, by a local computer program executed by client system, a federated machine learning model from a federated model server;
testing, by the local computer program and using a policy service, the federated machine learning model for vulnerabilities to attacks;
accepting, by the local computer program, the federated machine learning model in response to the federated machine learning model passing the testing;
training, by the local computer program, the federated machine learning model using input data comprising local data and outputting training parameters;
identifying, by the local computer program using the policy service, contamination by comparing the training parameters to the input data using an inversion of gradients of the training parameters to the input data; and
providing, by the local computer program, the training parameters to the federated model server.

2. The method of claim 1 , wherein the federated machine learning model is tested for vulnerabilities to attacks using brute force trials.

3. The method of claim 1 , wherein the federated machine learning model is tested for vulnerabilities to attacks using numerical simulation.

4. The method of claim 1 , wherein the comparing uses correlation tests to correlate the training parameters to the input data.

5. The method of claim 1 , further comprising:
rejecting, by the local computer program, the federated machine learning model in response to an identification of contamination.

6. The method of claim 1 , further comprising:
adding, by the local computer program, noise to the training parameters in response to an identification of contamination.

7. The method of claim 1 , further comprising:
executing, by the local computer program, a plurality of runs using the federated machine learning model before sending the training parameters to the federated model server.

8. A non-transitory computer readable storage medium, including instructions stored thereon, which when read and executed by one or more computer processors, cause the one or more computer processors to perform steps comprising:
receiving a federated machine learning model from a federated model server;
testing, using a policy service, the federated machine learning model for vulnerabilities to attacks;
accepting the federated machine learning model in response to the federated machine learning model passing the testing;
training the federated machine learning model using input data comprising local data and outputting training parameters;
identifying, using the policy service, accidental leakage by comparing the training parameters to the input data using an inversion of gradients of the training parameters to the input data; and
providing the training parameters to the federated model server.

9. The non-transitory computer readable storage medium of claim 8 , wherein the federated machine learning model is tested for vulnerabilities to attacks using brute force trials or using numerical simulation.

10. The non-transitory computer readable storage medium of claim 8 , further comprising instructions stored thereon, which when read and executed by one or more computer processors, cause the one or more computer processors to perform steps comprising adding noise to the training parameters in response to an identification of accidental leakage.

Assignees

• Jpmorgan Chase Bank Na

Inventors

• Eloul Shaltiel
• Moran Sean
• Silavong Fanny
• Kamthe Sanket
• Georgiadis Antonios

Classifications

• G06N20/00

  Machine learning · CPC title

• G06F2221/033

  Test or assess software · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06N3/098

  Distributed learning, e.g. federated learning · CPC title

• G06F21/577 Primary

  Assessing vulnerabilities and evaluating computer system security · CPC title