Description
Learn the concept of Federated Learning
Understand decentralized federated optimization techniques
Describe Federated Learning architectures
Elaborate on Differential Privacy on Federated Learning
Understand Federated Learning utilization in IoT, Autonomous, and Edge Computing Systems
Develop Federated Learning programs for NLP and Computer Vision systems
Implement Transfer Learning and Reinforcement Learning mechanisms for Federated Learning
solutions
Apply adaptive resource allocation approaches in Federated Learning framework
Recognize the ethical considerations and challenges facing Federated Learning
Program
Artificial Intelligence 132 CRs
Objectives
- This course provides a comprehensive introduction to federated learning,
an innovative decentralized approach to machine learning that enables
collaborative training on distributed data while prioritizing privacy and
data security. The course covers a wide range of topics, both theoretically
and practically, including decentralized optimization, secure aggregation,
communication-efficient protocols, and privacy-preserving techniques.
By the end of the course, students will have a solid understanding of
popular approaches used in federated learning, major architectures employed in federated learning systems, and the ability to construct and
scale a simple federated system. Additionally, students will develop
intuition for related technologies such as differential privacy and secure
aggregation, allowing them to effectively implement these techniques
within typical federated learning settings. The course also addresses the
challenges and ethical considerations associated with federated learning,
enabling students to reason about the heterogeneity and decentralization
issues that arise in federated systems.
