Federated Learning Strategies Over Wireless Channels

P. 525–533.

Ali A.

Machine learning over distributed data collected by many clients has
important applications in use cases where data privacy is a key concern or central data storage is not an option. Federated learning has introduced solutions
for these scenarios, unlike the client-server approach, where all the training data
is centralized in the server side, the clients, in a federated learning approach,
perform machine learning updates locally over their data and the central server
merely aggregates the resulting models without accessing the client’s local data.
Article reviews the characteristics and learning objectives of the federated
learning setting and gives an overview of the base algorithms that have been
developed for FL, compare between FEDAVG and Consensus algorithm and
shows the drawbacks and advantages of Consensus algorithm.
 

Keywords: Consensus, Federated learning, FEDAVG, IID, non-IID, Distributed strategies

In book

22nd International Conference, NEW2AN 2022, Tashkent, Uzbekistan, December 15–16, 2022, Proceedings. Internet of Things, Smart Spaces, and Next Generation Networks and Systems. LNCS, volume 13772

Issue 13772. , Springer, 2023.