Homomorphic Encryption-Based Federated Privacy Preservation for Deep Active Learning
Active learning is a technique for maximizing performance of machine learning with minimal labeling effort and letting the machine automatically and adaptively select the most informative data for labeling. Since the labels on records may contain sensitive information, privacy-preserving mechanisms...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-10-01
|
| Series: | Entropy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1099-4300/24/11/1545 |
| _version_ | 1797468368000778240 |
|---|---|
| author | Hendra Kurniawan Masahiro Mambo |
| author_facet | Hendra Kurniawan Masahiro Mambo |
| author_sort | Hendra Kurniawan |
| collection | DOAJ |
| description | Active learning is a technique for maximizing performance of machine learning with minimal labeling effort and letting the machine automatically and adaptively select the most informative data for labeling. Since the labels on records may contain sensitive information, privacy-preserving mechanisms should be integrated into active learning. We propose a privacy-preservation scheme for active learning using homomorphic encryption-based federated learning. Federated learning provides distributed computation from multiple clients, and homomorphic encryption enhances the privacy preservation of user data with a strong security level. The experimental result shows that the proposed homomorphic encryption-based federated learning scheme can preserve privacy in active learning while maintaining model accuracy. Furthermore, we also provide a Deep Leakage Gradient comparison. The proposed scheme has no gradient leakage compared to the related schemes that have more than 74% gradient leakage. |
| first_indexed | 2024-03-09T19:05:29Z |
| format | Article |
| id | doaj.art-6f1d70946fcf41bdaf388b742235681c |
| institution | Directory Open Access Journal |
| issn | 1099-4300 |
| language | English |
| last_indexed | 2024-03-09T19:05:29Z |
| publishDate | 2022-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Entropy |
| spelling | doaj.art-6f1d70946fcf41bdaf388b742235681c2023-11-24T04:36:00ZengMDPI AGEntropy1099-43002022-10-012411154510.3390/e24111545Homomorphic Encryption-Based Federated Privacy Preservation for Deep Active LearningHendra Kurniawan0Masahiro Mambo1Graduate School of Natural Science and Technology, Kanazawa University, Kanazawa 920-1192, JapanInstitute of Science and Engineering, Kanazawa University, Kanazawa 920-1192, JapanActive learning is a technique for maximizing performance of machine learning with minimal labeling effort and letting the machine automatically and adaptively select the most informative data for labeling. Since the labels on records may contain sensitive information, privacy-preserving mechanisms should be integrated into active learning. We propose a privacy-preservation scheme for active learning using homomorphic encryption-based federated learning. Federated learning provides distributed computation from multiple clients, and homomorphic encryption enhances the privacy preservation of user data with a strong security level. The experimental result shows that the proposed homomorphic encryption-based federated learning scheme can preserve privacy in active learning while maintaining model accuracy. Furthermore, we also provide a Deep Leakage Gradient comparison. The proposed scheme has no gradient leakage compared to the related schemes that have more than 74% gradient leakage.https://www.mdpi.com/1099-4300/24/11/1545privacy preservingfederated learningactive learninghomomorphic encryption |
| spellingShingle | Hendra Kurniawan Masahiro Mambo Homomorphic Encryption-Based Federated Privacy Preservation for Deep Active Learning Entropy privacy preserving federated learning active learning homomorphic encryption |
| title | Homomorphic Encryption-Based Federated Privacy Preservation for Deep Active Learning |
| title_full | Homomorphic Encryption-Based Federated Privacy Preservation for Deep Active Learning |
| title_fullStr | Homomorphic Encryption-Based Federated Privacy Preservation for Deep Active Learning |
| title_full_unstemmed | Homomorphic Encryption-Based Federated Privacy Preservation for Deep Active Learning |
| title_short | Homomorphic Encryption-Based Federated Privacy Preservation for Deep Active Learning |
| title_sort | homomorphic encryption based federated privacy preservation for deep active learning |
| topic | privacy preserving federated learning active learning homomorphic encryption |
| url | https://www.mdpi.com/1099-4300/24/11/1545 |
| work_keys_str_mv | AT hendrakurniawan homomorphicencryptionbasedfederatedprivacypreservationfordeepactivelearning AT masahiromambo homomorphicencryptionbasedfederatedprivacypreservationfordeepactivelearning |