Federated learning is used in distributed collaborative networks where multiple clients coordinate to train AI models without the need to share raw data. This is advantageous in a number of settings; for example; in health care where there are privacy/ethical issues associated with sharing data across multiple sites and in IoT networks where the sharing of raw data is costly in terms of the power consumption associated with data transmission.
In healthcare settings, one of the challenges is that much of the existing work on Federated learning has the objective of creating the most globally accurate model possible. However, the model created may not be best suited as a recommender for each individual patient. A recent proposal to overcome this drawback is the use of personalised federated learning models.
In this project we aim to look at explore the use of these personalised federated learning models in healthcare settings; drawing on the large number of open access data sets available and exploring the practicalities of the use of federated learning across data sets drawn from multiple sources.
You need to be comfortable working with maths, networking and programming to undertake this project.