Please use this identifier to cite or link to this item:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/79662Full metadata record
| ???org.dspace.app.webui.jsptag.ItemTag.dcfield??? | Value | Language |
|---|---|---|
| dc.contributor.advisor | 廖婉君(Wanjiun Liao) | |
| dc.contributor.author | Hao-Wei Tseng | en |
| dc.contributor.author | 曾浩瑋 | zh_TW |
| dc.date.accessioned | 2022-11-23T09:06:46Z | - |
| dc.date.available | 2021-09-11 | |
| dc.date.available | 2022-11-23T09:06:46Z | - |
| dc.date.copyright | 2021-09-11 | |
| dc.date.issued | 2021 | |
| dc.date.submitted | 2021-09-03 | |
| dc.identifier.citation | [1] McMahan, Brendan, et al. 'Communication-efficient learning of deep networks from decentralized data.' Artificial Intelligence and Statistics. PMLR, 2017. [2] Zhu, Ligeng, and Song Han. 'Deep leakage from gradients.' Federated learning.Springer, Cham, 2020. 17-31. [3] Melis, Luca, et al. 'Exploiting unintended feature leakage in collaborative learning.' 2019 IEEE Symposium on Security and Privacy (SP). IEEE, 2019. [4] Dwork, Cynthia, et al. 'Calibrating noise to sensitivity in private data analysis.' Theory of cryptography conference. Springer, Berlin, Heidelberg, 2006. [5] Zhao, Yang, et al. 'Local differential privacy-based federated learning for internet of things.' IEEE Internet of Things Journal. 2020. [6] Bonawitz, Keith, et al. 'Practical secure aggregation for privacy-preserving machine learning.' proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security. 2017. [7] Bagdasaryan, Eugene, Omid Poursaeed, and Vitaly Shmatikov. 'Differential privacy has disparate impact on model accuracy.' NIPS. 2019. [8] Kairouz, Peter, et al. 'Advances and open problems in federated learning.' arXiv preprint arXiv:1912.04977 (2019). [9] Nasr, Milad, Reza Shokri, and Amir Houmansadr. 'Comprehensive privacy analysis of deep learning: Passive and active white-box inference attacks against centralized and federated learning.' 2019 IEEE symposium on security and privacy (SP). IEEE, 2019. [10] Pyrgelis, Apostolos, Carmela Troncoso, and Emiliano De Cristofaro. 'Knock knock, who's there? Membership inference on aggregate location data.' arXiv preprint arXiv:1708.06145 (2017) [11] Dwork, Cynthia, et al. 'Robust traceability from trace amounts.' 2015 IEEE 56th Annual Symposium on Foundations of Computer Science. IEEE, 2015. [12] Hitaj, Briland, Giuseppe Ateniese, and Fernando Perez-Cruz. 'Deep models under the GAN: information leakage from collaborative deep learning.' Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security. 2017. [13] Truex, Stacey, et al. 'A hybrid approach to privacy-preserving federated learning.' Proceedings of the 12th ACM Workshop on Artificial Intelligence and Security. 2019. [14] Shayan, Muhammad, et al. 'Biscotti: A ledger for private and secure peer-to-peer machine learning.' arXiv preprint arXiv:1811.09904 (2018). [15] Mugunthan, Vaikkunth, Ravi Rahman, and Lalana Kagal. 'BlockFLow: An Accountable and Privacy-Preserving Solution for Federated Learning.' arXiv preprint arXiv:2007.03856 (2020). [16] Girgis, Antonious M., et al. 'Shuffled Model of Federated Learning: Privacy, Communication and Accuracy Trade-offs.' arXiv preprint arXiv:2008.07180 (2020). [17] Zhang, Chengliang, et al. 'Batchcrypt: Efficient homomorphic encryption for cross-silo federated learning.” ATC. 2020. [18] Chase, Melissa, et al. 'Private Collaborative Neural Network Learning.' IACR Cryptol. ePrint Arch. 2017 (2017): 762. [19] Corrigan-Gibbs, Henry, and Dan Boneh. 'Prio: Private, robust, and scalable computation of aggregate statistics.' 14th {USENIX} Symposium on Networked Systems Design and Implementation ({NSDI} 17). 2017. [20] Y. Li, Y. Zhou, A. Jolfaei, D. Yu, G. Xu and X. Zheng, 'Privacy-Preserving Federated Learning Framework Based on Chained Secure Multiparty Computing,' in IEEE Internet of Things Journal, vol. 8, no. 8, pp. 6178-6186, 15 April15, 2021. [21] Kadhe, Swanand, et al. 'FastSecAgg: Scalable Secure Aggregation for Privacy-Preserving Federated Learning.' arXiv preprint arXiv:2009.11248 (2020). [22] Bittau, Andrea, et al. 'Prochlo: Strong privacy for analytics in the crowd.' Proceedings of the 26th Symposium on Operating Systems Principles. 2017. [23] alle, Borja, et al. 'Privacy amplification via random check-ins.' arXiv preprint arXiv:2007.06605 (2020). [24] Rivest, Ronald L., Adi Shamir, and Yael Tauman. 'How to leak a secret.' International Conference on the Theory and Application of Cryptology and Information Security. Springer, Berlin, Heidelberg, 2001. [25] Sasson, Eli Ben, et al. 'Zerocash: Decentralized anonymous payments from bitcoin.' 2014 IEEE Symposium on Security and Privacy. IEEE, 2014. [26] Ruffing, Tim, Pedro Moreno-Sanchez, and Aniket Kate. 'Coinshuffle: Practical decentralized coin mixing for bitcoin.' European Symposium on Research in Computer Security. Springer, Cham, 2014. [27] Mihir Bellare and Phillip Rogaway. Random oracles are practical: A paradigm for designing efficient protocols. In Proceedings of the 1st ACM conference on Computer and communications security, pages 62–73. ACM, 1993. [28] Neal Koblitz and Alfred J Menezes. The random oracle model: a twenty-year retrospective. Designs, Codes and Cryptography, 77(2-3):587–610, 2015. [29] Kourou, Konstantina, et al. 'Machine learning applications in cancer prognosis and prediction.' Computational and structural biotechnology journal 13 (2015): 8-17. [30] Bagdasaryan, Eugene, et al. 'How to backdoor federated learning.' International Conference on Artificial Intelligence and Statistics. PMLR, 2020. [31] M. Salehi and E. Hossain, 'Federated Learning in Unreliable and Resource-Constrained Cellular Wireless Networks,' in IEEE Transactions on Communications, vol. 69, no. 8, pp. 5136-5151, Aug. 2021 | |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/79662 | - |
| dc.description.abstract | PriDA為確保隱私的安全聚合系統，於聯邦學習中計算聚合模型時同時保護每一位參與者之資料免於洩漏。只要參與者中存在一位非惡意攻擊者，任何參與者之資料皆能在安全聚合的過程中受保護。本文利用去中心化匿名以及資料混淆使得惡意攻擊者僅能獲得聚合過後之模型而不是特定參與者之私密資料。擴展了原始安全聚合之限制，本文透過動態聚合者選擇避免單點攻擊，並且透過去中心化匿名以及資料混淆放鬆了原先安全聚合需要多數參與者皆為誠實之參與者之限制，而是於不需要第三方信任機構存在之下，僅需要一位參與者為誠實之參與者即能保證每一位誠實參與者之隱私。 | zh_TW |
| dc.description.provenance | Made available in DSpace on 2022-11-23T09:06:46Z (GMT). No. of bitstreams: 1 U0001-0109202123165200.pdf: 750269 bytes, checksum: 209278043b20b5cf2103bfc802d9f469 (MD5) Previous issue date: 2021 | en |
| dc.description.tableofcontents | 摘要 i Abstract ii List of Figures iii List of Tables iii 1 Introduction 1 1.1 Background 1 1.2 Related Works 3 1.3 Multiparty computation approaches in federated learning 5 1.4 Preliminaries 6 1.4.1 Federated Learning 6 1.4.2 Data Obfuscation 6 1.4.3 Decentralized Anonymous 7 1.5 Thesis Organization 8 2 System model 9 2.1 Threat model 10 2.2 System model 11 3 Protocol 14 3.1 Setup 14 3.2 Aggregators Selection 15 3.3 Decentralized anonymous masking 16 3.4 Aggregation 17 4 Security Analysis and Evaluation 20 4.1 Security Proof 20 4.1.1 Honest-but-Curious Clients 20 4.1.2 Honest-but-Curious Clients and Aggregators 21 4.1.3 Malicious Clients 21 4.1.4 Malicious Clients and Aggregators 22 4.2 Performance Evaluation 24 5 Conclusion 27 Bibliography 28 | |
| dc.language.iso | en | |
| dc.title | PriDA:隱私維護去中心化匿名安全聚合於聯邦學習 | zh_TW |
| dc.title | PriDA: Privacy Preserving Decentralized Anonymous Secure Aggregation in Federated Learning | en |
| dc.date.schoolyear | 109-2 | |
| dc.description.degree | 碩士 | |
| dc.contributor.oralexamcommittee | 郭耀煌(Hsin-Tsai Liu),黃彥男(Chih-Yang Tseng),楊柏因 | |
| dc.subject.keyword | 聯邦學習,安全聚合,去中心化匿名, | zh_TW |
| dc.subject.keyword | federated learning,secure aggregation,decentralized anonymous, | en |
| dc.relation.page | 31 | |
| dc.identifier.doi | 10.6342/NTU202102936 | |
| dc.rights.note | 同意授權(全球公開) | |
| dc.date.accepted | 2021-09-03 | |
| dc.contributor.author-college | 電機資訊學院 | zh_TW |
| dc.contributor.author-dept | 電機工程學研究所 | zh_TW |
| Appears in Collections: | 電機工程學系 | |
Files in This Item:
| File | Size | Format | |
|---|---|---|---|
| U0001-0109202123165200.pdf | 732.68 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.