Accuracy Degrading: Toward Participation-Fair Federated Learning

Review Computer Science, Information Systems

Challenges and future directions of secure federated learning: a survey

Kaiyue Zhang et al.

Summary: As privacy concerns grow, federated learning has emerged to protect users' sensitive information. However, due to its distributed nature, federated learning is more vulnerable to attacks and there is still room for improvement in the current frameworks.

FRONTIERS OF COMPUTER SCIENCE (2022)

Add to Collection

Article Computer Science, Information Systems

Reputation-Based Federated Learning for Secure Wireless Networks

Zhendong Song et al.

Summary: Federated learning (FL) is a distributed machine learning method that protects data privacy by distributing and aggregating model parameters. The proposed reputation-based scheduling policy can effectively distinguish and defend against malicious user attacks. The study found that an increase in the percentage of malicious user devices has a negative impact on the convergence performance of FL.

IEEE INTERNET OF THINGS JOURNAL (2022)

Add to Collection

Proceedings Paper Acoustics

FEDERATED LEARNING CHALLENGES AND OPPORTUNITIES: AN OUTLOOK

Jie Ding et al.

Summary: This paper provides an outlook on the development of federated learning (FL) and highlights five emerging directions, including algorithm foundation, personalization, hardware and security constraints, lifelong learning, and nonstandard data. The perspectives shared in this paper are based on practical observations from large-scale federated systems for edge devices.

2022 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP) (2022)

Add to Collection

Proceedings Paper Computer Science, Artificial Intelligence

Incentive Mechanism for Horizontal Federated Learning Based on Reputation and Reverse Auction

Jingwen Zhang et al.

Summary: Current research on federated learning mainly focuses on joint optimization, improving efficiency and effectiveness, as well as protecting privacy, with relatively few studies on incentive mechanisms. This paper proposes an incentive mechanism based on reputation and reverse auction theory, demonstrating its effectiveness by selecting and rewarding participants through bidding for tasks and considering reputation.

PROCEEDINGS OF THE WORLD WIDE WEB CONFERENCE 2021 (WWW 2021) (2021)

Add to Collection

Article Telecommunications

Deep learning for the internet of things: Potential benefits and use-cases

Tausifa Jan Saleem et al.

Summary: The deployment of numerous sensors in the Internet of Things generates massive amounts of data, with DL being crucial in extracting valuable insights and enhancing IoT. The paper introduces DNN and its various architectures, discusses the potential benefits of DL in IoT, and presents a DL-based HAR model for performance comparison with other machine learning techniques.

DIGITAL COMMUNICATIONS AND NETWORKS (2021)

Add to Collection

Article Engineering, Electrical & Electronic