IP Library › Granted Patent US 12,632,795
Granted Patent B2
US 12,632,795 · App. 17/687,074 · Granted May 19, 2026

Data subset selection for federated learning

Inventors: Dinesh C. Verma (New Castle, NY); Supriyo Chakraborty (White Plains, NY); Shiqiang Wang (White Plains, NY); Augusto Vega (Poway, CA); Hazar Yueksel (San Francisco, CA); Ashish Verma (Nanuet, NY); Pradip Bose (Yorktown Heights, NY); Jayaram Kallapalayam Radhakrishnan (Pleasantville, NY)
Assignee: International Business Machines Corporation
G06N20/20G06N3/098G06N3/045
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,632,795
App. No.
17/687,074
Granted
May 19, 2026
Kind
B2
Abstract

Second machine learning models trained using respective second data sets can be received. The second machine learning models can be run using a first data set used in training a first machine learning model, where the second machine learning models produce respective outputs. Scores associated with the second machine learning models can be determined by comparing the respective outputs with ground truth associated with the first data set. Based on the scores associated with the second machine learning models, whether the first data set is to be discarded or kept can be determined for training the first machine learning model.

Claims (41)

1 . A system comprising:

a processor; and

a memory device coupled with the processor;

the processor configured to:

receive second machine learning models trained using respective second data sets;

run the second machine learning models using a first data set which was used in training a first machine learning model, the second machine learning models producing respective outputs;

determine scores associated with the second machine learning models by comparing the respective outputs with ground truth associated with the first data set, wherein each of the scores indicates how often a respective second machine learning model provides a right prediction using the first data set, in predicting an outcome the second machine learning models and the first machine learning model have been trained to predict;

based on the scores associated with the second machine learning models, determine which training data in the first data set is to be discarded or kept for training the first machine learning model,

wherein the first machine learning model checks the first data set using each of the second machine learning models, and each of the first data set and the respective second data sets has different data quality; and

sharing the scores with processors that run the second machine learning models without sharing the first data set used in training the first machine learning model with the second machine learning models.

2 . The system of claim 1 , wherein the first machine learning model and the second machine learning models are distributed across different computer systems and the first machine learning model is trained independently from the second machine learning models, wherein the first data set and the respective second data sets are unshared among the different computer systems.

3 . The system of claim 1 , wherein an ensemble model is built based on at least the scores associated with the second machine learning models, wherein which training data in the first data set is to be discarded or kept for training the first machine learning model is determined by running the ensemble model using the first data set and comparing an output of the ensemble model with the ground truth associated with the first data set.

4 . The system of claim 1 , wherein the processor is further configured to receive an ensemble model built based on at least the scores associated with the second machine learning models, run the ensemble model using the first data set, and select the training data from the first data set having ground truth that matches with an output of the ensemble model.

5 . The system of claim 1 , wherein the second machine learning models are weighted based on the scores associated with the second machine learning models.

6 . The system of claim 1 , wherein the processor is further configured to train the first machine learning model using the training data in the first data set determined to be kept.

7 . The system of claim 1 , wherein the processor is further configured to exchange the first machine learning model trained using the first data set with a plurality of sites running the second machine learning models.

8 . A computer-implemented method comprising:

receiving second machine learning models trained using respective second data sets;

running the second machine learning models using a first data set which was used in training a first machine learning model, the second machine learning models producing respective outputs;

determining scores associated with the second machine learning models by comparing the respective outputs with ground truth associated with the first data set, wherein each of the scores indicates how often a respective second machine learning model provides a right prediction using the first data set, in predicting an outcome the second machine learning models and the first machine learning model have been trained to predict;

based on the scores associated with the second machine learning models, determining which training data in the first data set is to be discarded or kept for training the first machine learning model,

wherein the first machine learning model checks the first data set using each of the second machine learning models, and each of the first data set and the respective second data sets has different data quality; and

sharing the scores with processors that run the second machine learning models without sharing the first data set used in training the first machine learning model with the second machine learning models.

9 . The computer-implemented method of claim 8 , wherein the first machine learning model and the second machine learning models are distributed across different computer systems and the first machine learning model is trained independently from the second machine learning models, wherein the first data set and the respective second data sets are unshared among the different computer systems.

10 . The computer-implemented method of claim 8 , wherein an ensemble model is built based on at least the scores associated with the second machine learning models, wherein which training data in the first data set is to be discarded or kept for training the first machine learning model is determined by running the ensemble model using the first data set and comparing an output of the ensemble model with the ground truth associated with the first data set.

11 . The computer-implemented method of claim 8 , wherein the method further includes receiving an ensemble model built based on at least the scores associated with the second machine learning models, running the ensemble model using the first data set, and selecting the training data in the first data set having ground truth that matches with an output of the ensemble model.

12 . The computer-implemented method of claim 8 , wherein the second machine learning models are weighted based on the scores associated with the second machine learning models.

13 . The computer-implemented method of claim 8 , wherein the method further includes training the first machine learning model using the training data in first data set determined to be kept.

14 . The computer-implemented method of claim 8 , wherein the method further includes exchanging the first machine learning model trained using the first data set with a plurality of sites respectively running the second machine learning models.

15 . A computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions readable by a device to cause the device to:

receive second machine learning models trained using respective second data sets;

run the second machine learning models using a first data set which was used in training a first machine learning model, the second machine learning models producing respective outputs;

determine scores associated with the second machine learning models by comparing the respective outputs with ground truth associated with the first data set, wherein each of the scores indicates how often a respective second machine learning model provides a right prediction using the first data set, in predicting an outcome the second machine learning models and the first machine learning model have been trained to predict;

based on the scores associated with the second machine learning models, determine which training data in the first data set is to be discarded or kept for training the first machine learning model,

wherein the first machine learning model checks the first data set using each of the second machine learning models, and each of the first data set and the respective second data sets has different data quality; and

sharing the scores with processors that run the second machine learning models without sharing the first data set used in training the first machine learning model with the second machine learning models.

16 . The computer program product of claim 15 , wherein the first machine learning model and the second machine learning models are distributed across different computer systems and the first machine learning model is trained independently from the second machine learning models, wherein the first data set and the respective second data sets are unshared among the different computer systems.

17 . The computer program product of claim 15 , wherein an ensemble model is built based on at least the scores associated with the second machine learning models, wherein which training data in the first data set is to be discarded or kept for training the first machine learning model is determined by running the ensemble model using the first data set and comparing an output of the ensemble model with the ground truth associated with the first data set.

18 . The computer program product of claim 15 , wherein the device is further caused to receive an ensemble model built based on at least the scores associated with the second machine learning models, run the ensemble model using the first data set, and select the training data from the first data set having ground truth that matches with an output of the ensemble model.

19 . The computer program product of claim 15 , wherein the second machine learning models are weighted based on the scores associated with the second machine learning models.

20 . The computer program product of claim 15 , wherein the device is further caused to train the first machine learning model using the training data in the first data set determined to be kept, and to exchange the first machine learning model trained with a plurality of sites respectively running the second machine learning models.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 4, 2022
From: VERMA, DINESH C.; CHAKRABORTY, SUPRIYO; WANG, SHIQIANG; VEGA, AUGUSTO; YUEKSEL, HAZAR; VERMA, ASHISH; BOSE, PRADIP; RADHAKRISHNAN, JAYARAM KALLAPALAYAM
To: INTERNATIONAL BUSINESS MACHINES CORPORATION
Reel/Frame 059175/0404 →
Continuity (1)
Related Publication 20230281518A1 · Sep 7, 2023
References Cited (49)
US 10341374B1 · Sadaghiani · 2019 [cited by examiner]
US 10775977B2 · Sowden et al. · 2020 [cited by applicant]
US 10970402B2 · Verma et al. · 2021 [cited by applicant]
US 20200311615A1 · Jammalamadaka · 2020 [cited by examiner]
US 20210073678A1 · Chu et al. · 2021 [cited by applicant]
US 20210409197A1 · Baracaldo et al. · 2021 [cited by applicant]
US 20220414464A1 · Krishnaswamy · 2022 [cited by examiner]
US 20230041015A1 · Chhibber · 2023 [cited by examiner]
US 20230259812A1 · Jagyasi et al. · 2023 [cited by applicant]
US 20230396434A1 · Liu et al. · 2023 [cited by applicant]
US 20240005216A1 · Radhakrishnan et al. · 2024 [cited by applicant]
CN 110995737A · 2020 [cited by applicant]
CN 111241580A · 2020 [cited by applicant]
CN 112580821A · 2021 [cited by applicant]
CN 113837761A · 2021 [cited by applicant]
EP 4083868A1 · 2022 [cited by applicant]
WO 2021082647A1 · 2021 [cited by applicant]
WO 2021112831A1 · 2021 [cited by applicant]
Bonawitz et al., “Practical Secure Aggregation for Privacy-Preserving Machine Learning,” Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security, Oct.-Nov. 2017, pp. 1175-1191. [cited by applicant]
Confidential Computing Consortium. “A Technical Analysis of Confidential Computing.” Oct. 2021, v1.2. 18 pages. [cited by applicant]
Geiping et al. “Inverting Gradients—How easy is it to break privacy in federated learning?. ” Advances in Neural Information Processing Systems 33 (2020): 16937-16947. [cited by applicant]
Kaissis et al. “End-to-end privacy preserving deep learning on multi-institutional medical imaging.” Nature Machine Intelligence 3.6 (2021): 473-484. [cited by applicant]
Lai et al. “Oort: Efficient federated learning via guided participant selection.” 15th {USENIX} Symposium on Operating Systems Design and Implementation ({OSDI} 21). 2021. 18 pages. [cited by applicant]
Melis et al. “Exploiting unintended feature leakage in collaborative learning.” 2019 IEEE symposium on security and privacy (SP). IEEE, 2019. pp. 691-706. [cited by applicant]
Yang et al. “Federated Machine Learning: Concept and Applications.” arXiv preprint arXiv:1902.04885 (2019): pp. 1-19. [cited by applicant]
Zhao et al. “iDLG: Improved deep leakage from gradients.” arXiv preprint arXiv:2001.02610 (2020). 5 pages. [cited by applicant]
Zhu et al. “Deep leakage from gradients.” 33rd Conference on Neural Information Processing Systems (NeurIPS 2019). 11 pages. [cited by applicant]
Anonymously, “Method and System for Reducing Communication Cost and Improving Local Model Accuracy in Federated Learning”, An IP.com Prior Art Database Technical Disclosure, IP.com No. IPCOM000265654D, May 4, 2021, 6 pa… [cited by applicant]
Tuor, T., et al., “Overcoming Noisy and Irrelevant Data in Federated Learning”, arXiv:2001.08300v2, Jun. 23, 2020, 8 pages. [cited by applicant]
Tuor, T., et al., “Data Selection for Federated Learning with Relevant and Irrelevant Data at Clients”, arXiv:2001.08300v1, Jan. 22, 2020, 14 pages. [cited by applicant]
Lai, F., et al., “Oort: Efficient Federated Learning via Guided Participant Selection”, 2020, 17 pages. [cited by applicant]
Bonawitz, K., et al., “Towards Federated Learning at Scale: System Design”, Proceedings of the 2nd SysML Conference, Mar. 2019, 15 pages. [cited by applicant]
Bhagoji, A.N., et al., “Analyzing Federated Learning through an Adversarial Lens”, Proceedings of the 36 th International Conference on Machine Learning, Jan. 2018, PMLR 97, 2019, 10 pages. [cited by applicant]
Grueneberg, K., et al., “A Policy-based Approach for Measuring Data Quality”, 2019 IEEE International Conference on Big Data (Big Data), Dec. 2019, pp. 4026-4031. [cited by applicant]
Kang, J., et al., “Reliable Federated Learning for Mobile Networks”, arXiv:1910.06837v1, Oct. 14, 2019, 8 pages. [cited by applicant]
Kang, J., et al., “Incentive Mechanism for Reliable Federated Learning: A Joint Optimization Approach to Combining Reputation and Contract Theory”, IEEE Internet of Things Journal, Dec. 2019, 14 pages, vol. 6, Issue 6. [cited by applicant]
Verma, D., et al., “Self Generating Policies for Machine Learning in Coalition Environments”, Policy-Based Autonomic Data Governance, 2019, pp. 42-65. [cited by applicant]
Verma, D., et al., “Managing Training Data from Untrusted Partners Using Self-generating Policies”, Artificial Intelligence and Machine Learning for Multi-Domain Operations Applications, International Society for Optics… [cited by applicant]
NIST, “NIST Cloud Computing Program”, http://csrc.nist.gov/groups/SNS/cloud-computing/index.html, Created Dec. 1, 2016, Updated Oct. 6, 2017, 9 pages. [cited by applicant]
Deng, et al., AUCTION: Automated and Quality-Aware Client Selection Framework for Efficient Federated Learning, IEEE Transactions On Parallel And Distributed Systems, Aug. 2022, vol. 33, No. 8, pp. 1996-2009. [cited by applicant]
Dong, et al., Offloading Federated Learning Task to Edge Computing with Trust Execution Environment, IEEE 17th International Conference on Mobile Ad Hoc and Sensor Systems (MASS), Dec. 10-13, 2020, pp. 491-496. [cited by applicant]
Fan, et al., Improving Fairness for Data Valuation in Horizontal Federated Learning, IEEE 38th International Conference on Data Engineering (ICDE), May 2022, pp. 2440-2453. [cited by applicant]
Kanagavelu, et al., Two-Phase Multi-Party Computation Enabled Privacy-Preserving Federated Learning, arXiv:2005.11901v1 [cs.DC], May 25, 2020, 11 pages. [cited by applicant]
Mondal, et al., Flatee: Federated Learning Across Trusted Execution Environments, arXiv:2111.06867v1 [cs. CR], Nov. 12, 2021, 8 pages. [cited by applicant]
Wang, et al., Measure Contribution of Participants in Federated Learning, arXiv: 1909.08525v1 [cs.LG], Sep. 17, 2019, 8 pages. [cited by applicant]
Zhao, et al., Efficient Client Contribution Evaluation for Horizontal Federated Learning, ICASSP 2021—2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Feb. 26, 2021, 5 pages. [cited by applicant]
Kairouz et al. “Advances and Open Problems in Federated Learning”, arXiv: 1912.04977v1 [cs.LG], Dec. 10, 2019, 105 pages. [cited by applicant]
Kuznetsov et al. “SecureFL: Privacy Preserving Federated Learning with SGX and TrustZone” 2021 IEEE/ACM Symposium on Edge Computing (SEC), Dec. 2021, 13 pages. [cited by applicant]
No Author. “Confidential Computing: Hardware-Based Trusted Execution for Applications and Data”, Confidential Computing Consortium, Jan. 2021,15 pages. [cited by applicant]