This textbook fills the vacuum for a senior-level academic textbook for post-GPT AI. It equips students with the mathematical underpinnings needed to effectively use deep learning and generative learning techniques in the real world. The author provides a comprehensive view of neural networks, from data-driven ML to generative pre-trained transformers, accompanied by examples and case studies that illustrate real-world solutions. Targeted to students and researchers who will eventually become part of the AI workforce, each chapter contains problem sets that help students develop insight when applying deep learning techniques, making it of great use to students and scholars alike.

S.Y. Kung, an IEEE Life Fellow, is a Professor of Electrical and Computer Engineering at Princeton University. For fifty years, his research has focused on fields intersecting hardware architecture and artificial intelligence, including VLSI array processors, deep learning architectures, AI algorithms, and machine learning systems. Since 1990, he has been the founding Editor-in-Chief of the Journal of VLSI Signal Processing Systems. He was a recipient of the IEEE Third Millennium Medal in 2000 and the Distinguished Achievement Award from the Chinese Institute of Engineers (CIE-USA) in 2023. He is the author of several textbooks which provides the theoretical foundation for this book, including VLSI Array Processors (Prentice-Hall, 1988), Digital Neural Networks (Prentice-Hall, 1994), and Kernel Methods and Machine Learning (Cambridge University Press, 2014).