Design and Implementation of a Privacy Auditing Component for the Decentralized Federated Learning Framework

The decentralized nature of Federated Learning (FL) has led to its increased popularity in model training [1] [2]. However, it is crucial to prioritize privacy protection. This project aims to develop a privacy auditing component for a decentralized federated learning framework. Through the monitoring of mechanisms that preserve privacy, this component will enhance the security of data, ensure compliance with privacy regulations, and foster trust in the FL system [3] [4].

The main objectives of this project are as follows:

a) Privacy Audit Design: Develop a detailed design for the privacy auditing component that covers the various stages of the decentralized federated learning process, including data aggregation, model updates, and communication protocols.

b) Privacy Metrics: Define privacy metrics and evaluation criteria that will be used to assess the level of privacy preservation in the federated learning framework.

c) Implementation: Implement the privacy auditing component, integrating it seamlessly into the existing decentralized federated learning framework. Ensure that the auditing component can effectively track and analyze privacy-related events.

d) Auditing Dashboard: Create a user-friendly auditing dashboard that provides real-time insights into the privacy status of the federated learning system. This dashboard will enable administrators to monitor privacy metrics and receive alerts for potential privacy breaches.

e) Privacy Enhancement Recommendations: Based on the audit results, provide recommendations for improving the privacy-preserving mechanisms within the federated learning framework.

[1] Yin, X., Zhu, Y., & Hu, J. (2021). A comprehensive survey of privacy-preserving federated learning: A taxonomy, review, and future directions. ACM Computing Surveys (CSUR), 54(6), 1-36.

[2] Beltrán, E. T. M., Pérez, M. Q., Sánchez, P. M. S., Bernal, S. L., Bovet, G., Pérez, M. G., ... & Celdrán, A. H. (2022). Decentralized federated learning: Fundamentals, state-of-the-art, frameworks, trends, and challenges. arXiv preprint arXiv:2211.08413.

[3] https://github.com/enriquetomasmb/fedstellar

[4] Beltrán, E. T. M., Gómez, Á. L. P., Feng, C., Sánchez, P. M. S., Bernal, S. L., Bovet, G., ... & Celdrán, A. H. (2023). Fedstellar: A Platform for Decentralized Federated Learning. arXiv preprint arXiv:2306.09750.