Federated learning for edge computing

Abstract

This research paper goals to optimize federated learning by focusing on two key objectives. First, we'll evaluate how different traditional machine learning models perform when implementing federated learning strategies. I will test these models’ using data from a variety of sources to ensure they operate well in real-world scenarios while maintaining people's privacy. Second, I would like to understand more about the unique characteristics of decentralized datasets. By examining data from a variety of sources, I have attempted to determine what separates each dataset. This will allow me to create machine learning models that perform better on each dataset. I will also be looking at a variety of areas of federated learning, such as data partitioning, how to apply them, and the kind of data I will be working with. My aim is to discover the best ways to employ federated learning for a variety of data by evaluating various models and strategies. This research will help us better understand federated learning and develop more effective machine learning models for various situations. Finally, this will allow us to make greater use of data while maintaining people's privacy.