Federated Learning for Privacy-Preserving Edge Intelligence: Communication, Heterogeneity, and Security Challenges
Keywords:
Federated learning, edge intelligence, distributed machine learningAbstract
Federated learning emerged during the second half of the 2010s as a distributed machine-learning paradigm capable of training shared predictive models without requiring participating devices or organizations to centralize their raw data. This property makes federated learning particularly relevant to smartphones, Internet of Things devices, healthcare systems, financial institutions, and other environments in which privacy, bandwidth, and data-governance constraints limit conventional cloud-based learning. This review examines the technological foundations and principal research challenges of federated learning as understood through 2020. The Federated Averaging framework is discussed together with communication-efficient optimization, local computation, client selection, secure aggregation, differential privacy, heterogeneous device capabilities, and non-independent and identically distributed data. Particular attention is given to communication overhead, client dropout, variable computational resources, statistical heterogeneity, model convergence, and exposure of sensitive information through gradients or parameter updates. Developments such as FedProx are considered in relation to heterogeneous federated environments, while secure aggregation and differential privacy are evaluated as complementary privacy mechanisms. The review also discusses early applications to mobile keyboard prediction, healthcare, and edge intelligence. It concludes that federated learning represents a significant shift from centralized machine learning, but effective deployment requires simultaneous consideration of optimization, communication, privacy, security, and system heterogeneity.
References
McMahan HB, Moore E, Ramage D, Hampson S, Agüera y Arcas B. Communication-efficient learning of deep networks from decentralized data. Proc Mach Learn Res. 2017;54:1273-1282.
Konečný J, McMahan HB, Yu FX, Richtárik P, Suresh AT, Bacon D. Federated learning: strategies for improving communication efficiency. arXiv. 2016;1610.05492.
Bonawitz K, Ivanov V, Kreuter B, Marcedone A, McMahan HB, Patel S, et al. Practical secure aggregation for privacy-preserving machine learning. In: Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security. New York: ACM; 2017. p. 1175-1191.
Smith V, Chiang CK, Sanjabi M, Talwalkar AS. Federated multi-task learning. Adv Neural Inf Process Syst. 2017;30:4424-4434.
Yang Q, Liu Y, Chen T, Tong Y. Federated machine learning: concept and applications. ACM Trans Intell Syst Technol. 2019;10(2):12.
Kairouz P, McMahan HB, Avent B, Bellet A, Bennis M, Bhagoji AN, et al. Advances and open problems in federated learning. arXiv. 2019;1912.04977.
Li T, Sahu AK, Talwalkar A, Smith V. Federated learning: challenges, methods, and future directions. IEEE Signal Process Mag. 2020;37(3):50-60.
Li T, Sahu AK, Zaheer M, Sanjabi M, Talwalkar A, Smith V. Federated optimization in heterogeneous networks. Proc Mach Learn Syst. 2020;2:429-450.
Aledhari M, Razzak R, Parizi RM, Saeed F. Federated learning: a survey on enabling technologies, protocols, and applications. IEEE Access. 2020;8:140699-140725.
Rieke N, Hancox J, Li W, Milletarì F, Roth HR, Albarqouni S, et al. The future of digital health with federated learning. NPJ Digit Med. 2020;3:119.
Hard A, Rao K, Mathews R, Ramaswamy S, Beaufays F, Augenstein S, et al. Federated learning for mobile keyboard prediction. arXiv. 2018;1811.03604.
Caldas S, Duddu SMK, Wu P, Li T, Konečný J, McMahan HB, et al. LEAF: a benchmark for federated settings. arXiv. 2018;1812.01097.
Nishio T, Yonetani R. Client selection for federated learning with heterogeneous resources in mobile edge. In: IEEE International Conference on Communications. Piscataway: IEEE; 2019. p. 1-7.
Sattler F, Wiedemann S, Müller KR, Samek W. Robust and communication-efficient federated learning from non-IID data. IEEE Trans Neural Netw Learn Syst. 2020;31(9):3400-3413.
Zhao Y, Li M, Lai L, Suda N, Civin D, Chandra V. Federated learning with non-IID data. arXiv. 2018;1806.00582.
Zhu L, Liu Z, Han S. Deep leakage from gradients. Adv Neural Inf Process Syst. 2019;32:14774-14784.
Bagdasaryan E, Veit A, Hua Y, Estrin D, Shmatikov V. How to backdoor federated learning. Proc Mach Learn Res. 2020;108:2938-2948.
Geyer RC, Klein T, Nabi M. Differentially private federated learning: a client level perspective. arXiv. 2017;1712.07557.
Abadi M, Chu A, Goodfellow I, McMahan HB, Mironov I, Talwar K, Zhang L. Deep learning with differential privacy. In: Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security. New York: ACM; 2016. p. 308-318.
Shokri R, Shmatikov V. Privacy-preserving deep learning. In: Proceedings of the 22nd ACM SIGSAC Conference on Computer and Communications Security. New York: ACM; 2015. p. 1310-1321.
