IEEE SPS Seasonal School on Networked Federated Learning: Theory, Methods and Applications
Many important application domains generate distributed collections of local datasets. Networked federated learning allows to train tailored models for each local dataset in a col- laborative fashion.
This seasonal school teaches some of the theoretical and al- gorithmic underpinnings of federated learning. We illustrate key concepts using the toy example of a personalized Covid-19 diagnosis smartphone app.
This seasonal school is organised as three modules:
- Basics of Machine Learning
- Networked Data
- Networked Models
Each module consists of lectures and coding assignments with Python notebooks.
This seasonal school is inspired by the recent Live-Project by Alexander Jung, Assistant Professor for machine learning at the Department of Computer Science, Aalto University.