Author Description

"This is a unified, modern journey through information theory and machine learning! Information theory is the hidden mathematical language of today's AI systems. From classical statistics to deep learning, from representation learning to generative models, the principles of entropy, mutual information, and information flow shape how models learn, generalize, and reason. Information theory for machine learning provides a comprehensive and rigorous treatment of these ideas, bridging foundational theory with modern deep learning practice. Through clear explanations, illustrative diagrams, and hands-on Python implementations, this book reveals how information principles underlie PCA, clustering, variational autoencoders, normalizing flows, diffusion models, attention mechanisms, and large generative systems. Ideal for advanced undergraduate students, graduate students, researchers, and practitioners seeking a deeper understanding of the information-theoretic foundations of machine learning including the modern generative models that define today's AI." -- back cover.