Machine Learning in the Age of Artificial Intelligence Volume I

This volume is one part of a planned five-volume treatise on machine learning in the age of artificial
intelligence. It has been separated from the former collected first volume so that the reader may work through
a more reasonable physical and intellectual unit without losing the order of the original argument.
The division has not been made by accident of page count alone. A treatise should be cut where the joints are
anatomical, not where the printer grows weary. The first new volume establishes orientation, data,
mathematical foundations, and the first classical models. The second carries the classical arc into ensembles,
kernels, unsupervised methods, evaluation, and the first great architectures of deep learning. The later
volumes of the complete series are reserved for foundation models, agents, multimodal systems, governance,
domain practice, capstones, appendices, and professional reference.
The style remains deliberately discursive. A manual gives steps; a treatise gives reasons. Equations appear
when needed, and variables are introduced rather than lowered from a cloud with a small thunderclap. AI
assistants are treated as useful instruments under judgement, not as substitutes for the grammar of evidence.
The present volume contains Chapters 1-19. Its purpose is to preserve the original sequence while allowing
this part of the work to stand independently on the desk, in the hand, and in the mind.
The reader will find chapter presentation pages before each chapter, formal notation where the argument
requires it, worked examples, exercises with resolutions, references, and the sustained insistence that data,
models, metrics, and deployment are not separate provinces but a single disciplined chain.
The page numbers in the table below refer to the body of this new volume. Front matter is intentionally
unnumbered. Numbering begins with the presentation page of the first chapter in this part and continues
through the conclusion and index.