Journal Of Network security Review Article
Enhancing Data Security and Privacy in Machine Learning Using Federated
This is an unedited manuscript accepted for publication and provided as an Article in Press for early access at the author’s request. The article will undergo copyediting, typesetting, and galley proof review before final publication. Please be aware that errors may be identified during production that could affect the content. All legal disclaimers of the journal apply.
Abstract
Federated learning is a way of training machine learning models. It does this across devices or organizations without putting all the raw data in one place. This helps to keep data secure in modern artificial intelligence systems. This paper looks at how federated learning helps to keep data from being seen reduces the chance of something going wrong and lets people keep control over information. This makes the whole security system stronger for machine learning that people work on together.
We look at ways to keep data that work with federated learning. These include ways to combine data ways to keep data private and ways to keep data secret. We see how these methods affect how accurate the model is, how much data is sent and how much it costs to do the work. We also look at ways to make the system more secure like picking the devices making sure people want to participate and keeping the work safe.
We test these methods with life examples like healthcare and internet of things devices. We see how well these methods stop people from getting to the data turning the model around and poisoning the data. We also see what problems are still there with many layers of protection. Our results show that no one method can keep everything. Using methods together like secret codes, statistics and architecture works the best. This also means it takes more time and resources.
We talk about how hard it's to use these methods in many different situations and with devices that do not have many resources. It is also hard to make a way to measure threats in many different situations. finally, we give advice, on how to use learning in the real world and suggest what to research in the future. We think that using methods together making a standard way to measure threats and following data protection rules is important. All of this helps to make learning systems that are strong keep data private and can be used in real life especially in important areas.
Keywords
References (10)
- Kairouz P, McMahan HB. Advances and open problems in federated learning. Foundations and trends in machine learning. 2021 Jun 23;14(1-2):1-210.
- Dwork C, Roth A. The algorithmic foundations of differential privacy. Foundations and trends® in theoretical computer science. 2014 Aug 11;9(3-4):211-487.
- Bonawitz K, Ivanov V, Kreuter B, Marcedone A, McMahan HB, Patel S, Ramage D, Segal A, Seth K. Practical secure aggregation for privacy-preserving machine learning. Inproceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security 2017 Oct 30 (pp. 1175-1191).
- Li T, Sahu AK, Talwalkar A, Smith V. Federated learning: Challenges, methods, and future directions. IEEE signal processing magazine. 2020 May 1;37(3):50-60.
- McMahan B, Moore E, Ramage D, Hampson S, y Arcas BA. Communication-efficient learning of deep networks from decentralized data. InArtificial intelligence and statistics 2017 Apr 10 (pp. 1273-1282). Pmlr.
- Yang Q, Liu Y, Chen T, Tong Y. Federated machine learning: Concept and applications. ACM Transactions on Intelligent Systems and Technology (TIST). 2019 Jan 28;10(2):1-9.
- Mothukuri V, Parizi RM, Pouriyeh S, Huang Y, Dehghantanha A, Srivastava G. A survey on security and privacy of federated learning. Future Generation Computer Systems. 2021 Feb 1;115:619-40.
- Truex S, Baracaldo N, Anwar A, Steinke T, Ludwig H, Zhang R, Zhou Y. A hybrid approach to privacy-preserving federated learning. InProceedings of the 12th ACM workshop on artificial intelligence and security 2019 Nov 11 (pp. 1-11).
- Zhang C, Xie Y, Bai H, Yu B, Li W, Gao Y. A survey on federated learning. Knowledge-based systems. 2021 Mar 15;216:106775.
- Wei K, Li J, Ding M, Ma C, Yang HH, Farokhi F, Jin S, Quek TQ, Poor HV. Federated learning with differential privacy: Algorithms and performance analysis. IEEE transactions on information forensics and security. 2020 Apr 17;15:3454-69.