Information Theory for Data Transmission, Security and Machine Learning

Catalog description

This is a graduate-level introduction to mathematics of information theory. We will cover both classical and modern topics, starting from f-divergences, information measures and relations between them. With these tools we will study the fundamental limits of data transmission over noisy channels. Wiretap channels, where information-theoretic security versus a malicious eavesdropped must be ensured, will also be covered. Passive and adversarial models will be considered. Finally, we will explore connections between information theory and machine learning, examining how they can cross-fertilize each other.