Administrative Information
| Title | Federated Learning - Train deep models |
| Duration | 150 min |
| Module | C |
| Lesson Type | Practical |
| Focus | Technical - Future AI |
| Topic | Advances in ML models through a HC lens - A result Oriented Study |
Keywords
Federated Learning,Tensorflow,
Learning Goals
- Understand how to train models using the Federated Learning framework
- Understand how local data distribution affects the federated learning
- Becoming familiar with a high-level framework like TensorFlow
Expected Preparation
Learning Events to be Completed Before
Obligatory for Students
- Students should have a basic understanding of deep learning concepts and techniques
- Basic understanding of deep learning training (SGD, backpropagation algorithm) and evaluation techniques
Optional for Students
- A bit of knowledge of the TensorFlow framework and Python programming language
References and background for students
Recommended for Teachers
Lesson materials
Instructions for Teachers
- Provide a hands-on lecture where students can learn from guided exercises
- Propose pop-up quizzes
Outline
| Duration | Description | Concepts | Activity |
|---|---|---|---|
| 20 min | Introduction to the framework: how to code a simple federated learning system | Tools introduction | Introduction to main tools |
| 60 min | Federated Training: the easy way. How to apply train models with federated learning based on iid local data | Federated Average | Practical session and working examples |
| 60 min | Federated training: the hard way. How does heterogeneity affect Federated Average and what can we do | Challenges connected to Federated Learning | Practical session and working examples |
| 10 min | Conclusion, questions and answers | Summary | Conclusions |
Acknowledgements
The Human-Centered AI Masters programme was Co-Financed by the Connecting Europe Facility of the European Union Under Grant №CEF-TC-2020-1 Digital Skills 2020-EU-IA-0068.