Pushpendra Kumar Sikarwal,
Mukesh Kumar Gupta,
Adamya Gupta,
- Research scholar, Department of Computer Science and Engineering, Suresh Gyan Vihar University, Jaipur, Rajasthan, India
- Professor, Department of Electrical Engineering, Suresh Gyan Vihar University, Jaipur, Rajasthan, India
- Research scholar, Department of Computer Science and Engineering, Jaipur Engineering, College & Research Centre, Jaipur, Rajasthan, India
Abstract
The increasing deployment of parallel and distributed intelligent systems has intensified the need for privacy-preserving learning frameworks that can exploit multi-core and graphics processing unit (GPU)-based architectures without centralizing sensitive data. This work proposes a parallel adaptive federated learning (AFL) framework that integrates differential privacy and secure aggregation over heterogeneous multi-core and GPU platforms to enhance both data confidentiality and convergence efficiency. The framework dynamically adjusts client participation, learning rates, and aggregation weights across parallel clients to mitigate non-IID (non-independent and identically distributed) data effects and communication bottlenecks in large-scale interconnection environments. Extensive experiments on benchmark datasets, including CIFAR-10 and Modified National Institute of Standards and Technology (MNIST), demonstrate that the proposed parallel AFL configuration with differential privacy and secure aggregation achieves near-centralized accuracy while reducing convergence rounds and improving resistance to model inversion and gradient leakage attacks compared with conventional centralized and baseline federated schemes. These results show that coupling adaptive optimization with privacy-preserving mechanisms on multi-core and GPU-accelerated infrastructures offers a practical direction for scalable, secure, and high-performance parallel learning in sensitive domains.
Keywords: Adaptive federated learning, differential privacy, GPU accelerators, interconnection networks, multi-core architecture, parallel computing, privacy-preserving machine learning, secure aggregation
[This article belongs to Recent Trends in Parallel Computing ]
References
- Shawkat M, Ali ZH, Salem M, El-Desoky A. A robust and personalized privacy-preserving approach for adaptive clustered federated distillation.Sci Rep. 2025;15:14069. doi:10.1038/s41598-025-96468-8. PMID: 40269026.
- Shalabi E, Khedr W, Rushdy E, Salah A. A comparative study of privacy-preserving techniques in federated learning: a performance and security analysis. Information. 2025;16:244.doi:10.3390/info16030244.
- Thakur A, Sharma P, Clifton DA. Dynamic neural graphs based federated reptile for semi-supervised multi-tasking in healthcare applications. IEEE J Biomed Health Inform. 2022;26:1761–1772. doi:10.1109/JBHI.2021.3134835. PMID: 34898443.
- Chang Y, Zhang K, Gong J, Qian H. Privacy-preserving federated learning via functional encryption, revisited. IEEE Trans Inf Forensics Secur. 2023;18:1855–doi:10.1109/TIFS.2023 .3255171.
- He Z, Wang L, Cai Z. Clustered federated learning with adaptive local differential privacy on heterogeneous IoT data. IEEE Internet Things J. 2024;11:137–146. doi:10.1109/JIOT.2023.
- Li Y, Xu G, Meng X, Du W, Ren X. LF3PFL: a practical privacy-preserving federated learning algorithm based on local federalization scheme. Entropy (Basel). 2024;26:353. doi:10.3390/e26050353. PMID: 38785602.
- Jiao S, Cai L, Wang X, Cheng K, Gao X. A differential privacy federated learning scheme based on adaptive Gaussian noise. Comput Model Eng Sci. 2024;138(2):16791694. doi:10.32604/cmes.2023.030512.
- Batool H, Anjum A, Khan A, Izzo S, Mazzocca C, Jeon G. A secure and privacy preserved infrastructure for VANETs based on federated learning with local differential privacy. Inf Sci. 2024;652:119717. doi:10.1016/j.ins.2023.119717.
- Gamiz I, Regueiro C, Jacob E, Lage O, Higuero M. PRoT-FL: a privacy-preserving and robust training manager for federated learning. Inf Process Manag. 2025;62:103929.doi:10.1016/j.ipm.2024.103929.
- Chang Z, Liu S, Xiong X, Cai Z, Tu G. A survey of recent advances in edge-computing-powered artificial intelligence of things. IEEE Internet Things J. 2021;8:13849–13875. doi:10.1109/JIOT. 3088875.
- Wang Y, Yang S, Ren X, Zhao P, Zhao C, Yang X. IndustEdge: a time-sensitive networking enabled edge-cloud collaborative intelligent platform for smart industry. IEEE Trans Ind Inform. 2022;18:2386–2398. doi:10.1109/TII.2021.3104003.
- Tajabadi M, Martin R, Heider D. Privacy-preserving decentralized learning methods for biomedical applications. Comput Struct Biotechnol J. 2024;23:3281–3287. doi:10.1016/j.csbj.2024.08.024. PMID: 39296807.
- Anastasakis Z, Bourou S, Velivasaki TH, Voulkidis A, Skias D. Analysis of privacy preservation enhancements in federated learning frameworks. In: Sofia RC, Soldatos J, editors. Shaping the Future of IoT with Edge Intelligence: How Edge Computing Enables the Next Generation of IoT Applications. New York: River Publishers; 2024. p. 117–133. doi:10.1201/9781032632407-8.
- Zhan S, Huang L, Luo G, Zheng S, Gao Z, Chao HC. A review on federated learning architectures for privacy-preserving AI: lightweight and secure cloud-edge-end collaboration. Electronics. 2025;14:2512. doi:10.3390/electronics14132512.
- Xu R, Baracaldo N, Zhou Y, Anwar A, Kadhe S, Ludwig H. Detrust-FL: privacy-preserving federated learning in decentralized trust setting. 2022 IEEE 15th International Conference on Cloud Computing (CLOUD), Barcelona, Spain. 2022. p. 417–426. doi:10.1109/CLOUD55607.2022.00065.
Recent Trends in Parallel Computing
| Volume | 13 | |
| Issue | 01 | |
| Received | 01/02/2026 | |
| Accepted | 10/02/2026 | |
| Published | 20/03/2026 | |
| Publication Time | 47 Days |