Privacy-Preserving Mechanisms for Computer Algorithms 
and AI Collaborative Training Methods under the Federated
Learning Framework

Jian Zhang

doi:10.70711/aitr.v2i11.7424

Privacy-Preserving Mechanisms for Computer Algorithms and AI Collaborative Training Methods under the Federated Learning Framework

Jian Zhang

Abstract

Federated Learning (FL) arises as a key framework for tackling the substantial privacy risks of traditional centralized AI, supporting
collaborative modeling training on decentralized devices without sharing rawdata. This paper does a full analysis of the FL framework, looking at the main privacy preserving technologies like DP, HE, and SMPC, as well as more advanced collaborative training algorithms created to
solve key systemic problems such as the private information versus usefulness issue, inefficient communication, and non-statistically similar
data. Through this sort of structured analysis and comparison work, this research seeks to explain these various difficult trade-offs, and it ends
by giving an identification of open problems and future research directions for robust, privately conscious artificial intelligence.

Keywords

Federated Learning; Privacy Preservation Mechanisms; AI Collaborative Learning; Differential Privacy; Homomorphic Encryption; Security Multi-Party Computation; Data Privacy

Full Text:

PDF

Included Database

References

[1] Zareie S, Esmaeilyfard R, Shamsinejadbabaki P. A Privacy-Preserving Federated Learning Framework for Ambient Temperature Estimation With Crowdsensing and Exponential Mechanism[J]. International Journal of Intelligent Systems, 2025(1):5531568-5531568.

[2] Wang W L. Research on data security sharing and privacy protection mechanism based on federated learning[J]. Security & Informatization, 2025, (06):154-156.

[3] Guo Z Z, Tian Y L, Li M Q. Adaptive Privacy-Preserving Federated Learning Scheme Based on Incentive Mechanism[J/OL]. Computer

Engineering, 1-16[2025-07-14]. https://doi.org/10.19678/j.issn.1000-3428.0070751.

[4] Hu D Q, Zhang Z L, Kang Y, et al. Research on Privacy Protection Mechanism in Software Defect Prediction-Policy and Implementation under Federated Learning Framework[J]. Network Security Technology & Application, 2025, (05):28-30.

[5] Shenoy D, Bhat R, Prakasha K K. Exploring privacy mechanisms and metrics in federated learning[J]. Artificial Intelligence Review,

2025, 58(8):223-223.

[6] Zhao F, Yang B, Su Z, et al. A blockchain-enabled privacy-preserving and incentive mechanism-driven federated learning scheme for

IoV[J]. Computer Networks, 2025, 264111262-111262.

[7] Chen F, Li C H, Li X H, et al. Research on trusted computing of local model in federated learning based on model similarity[J]. Journal

of Guilin University of Electronic Technology, 2025, 45(01): 27-32. DOI:10.16725/j.1673-808X.2022259.

[8] Chen Z, Mo J B, Chen Q X, et al. Design and Optimization of Federated Learning Algorithms in Edge Computing Networks[J]. Mobile

Communications, 2024, 48(12):122-128.

[9] Miao J. algorithm optimization of device grouping and asynchronous federated learning based on NON-IID data[D]. Inner Mongolia

University, 2024. DOI:10.27224/d.cnki.gnmdu.2024.000145.

[10] Chen X Y. Research on energy efficiency optimization method of federated learning in edge computing environment[D]. Qingdao University, 2024. DOI:10.27262/d.cnki.gqdau.2024.002637.

[11] Li Y, Ge L, Jiang M. Fine-grained personalized federated learning via transformer in the transformer framework[J]. Knowledge-based

systems, 2024(Oct.9):112276.

DOI: http://dx.doi.org/10.70711/aitr.v2i11.7424

Refbacks

There are currently no refbacks.