[CFP] Federated Learning – Theoretical and Practical Advances

[Mirko Polato]

Frontiers in Big Data

Research Topic: Federated Learning – Theoretical and Practical Advances

Webpage: https://www.frontiersin.org/research-topics/27378/federated-learning---theoretical-and-practical-advances

DEADLINE

Abstract (not mandatory): 3 December 2021

Manuscript: 1 February 2022

About this Research Topic

Privacy and security are becoming key concerns in our digital age. Users are becoming aware of possible privacy issues, and they are increasingly reluctant in sharing their sensitive information with companies and organizations. Moreover, a series of new regulations, such as the European Union’s General Data Protection Regulation (GDPR), pose new legislative challenges to the big data and artificial intelligence (AI) community.

Federated learning (FL) is a new breed of AI that builds upon decentralized data and training that brings learning to the edge. This novel paradigm, proposed by Google, came to life mainly for three reasons: (1) Data protection using local data from edge devices instead of sending sensitive data to the server; (2) The unavailability of sufficient data to reside on the central server; (3) The need of secure and private-preserving techniques for fully decentralized/distributed learning.

Although still in its infancy, FL has already shown important theoretical and practical results making FL one of the hottest topics in the machine learning community. FL is one of the key enabling technologies of future intelligent applications in domains such as autonomous driving, smart manufacturing, healthcare, and many others.

This Research Topic aims to bring together solutions and technologies that help to realize the vision of future distributed learning applications through FL, provide the community with the current state of the art, and highlight the challenges. Potential topics include but are not limited to the following:

• Architecture and privacy-preserving learning protocols;
• Attacks to FL algorithm, e.g., adversarial learning, data poisoning, adversarial examples, adversarial robustness, black box attacks;
• Privacy-preserving techniques, e.g., secure multi-party computation, homomorphic encryption, secret sharing techniques, differential privacy;
• Decentralized Federated Learning, e.g., blockchain-based FL and gossip learning;
• Federated Transfer Learning algorithms;
• Horizontal/Vertical Federated Learning algorithms;
• Heterogeneous federated learning;
• Interesting applications for Federated Learning, e.g., in IoT, recommendation systems, medicine, sensor networks, and text processing.

Research Topic Editors

Dr. Mirko Polato, Department of Computer Science, Univerisity of Turin, Italy

Prof. Frank-Michael Schlief, University of Applied Sciences, Würzburg-Schweinfurt

Würzburg, Germany