Machine Learning

Machine Learning: A Constraint-Based Approach, 2nd Edition, authored by Marco Gori, Alessandro Betti, and Stefano Melacci, provides a comprehensive, mathematically grounded analysis of machine learning concepts. Published by Morgan Kaufmann in 2023, this 560-page computer science textbook establishes a unified framework that organizes machine learning models and computational algorithms around formal mathematical constraints.

The volume examines deep learning architectures and incorporates multivalued logic formalisms, such as fuzzy systems, into its theoretical framework. Across these diverse topics, the text develops a unified view of regularization, demonstrating how constraint-based regularization operators connect directly to the underlying parsimony principle.

To guide students through theoretical problem solving, all included solved exercises are categorized according to Donald Knuth's difficulty ranking scale. This explicit structure offers target support for upper-level and graduate students enrolled in computer science machine learning courses, while providing clear value for working professionals in artificial intelligence.