Machine Learning: A Constraint-Based Approach, Second Edition provides readers with a refreshing look at the basic models and algorithms of machine learning, with an emphasis on current topics of interest that include neural networks and kernel machines. The book presents the information in a truly unified manner that is based on the notion of learning from environmental constraints. It draws a path towards deep integration with machine learning that relies on the idea of adopting multivalued logic formalisms, such as in fuzzy systems. Special attention is given to deep learning, which nicely fits the constrained-based approach followed in this book.
The book presents a simpler unified notion of regularization, which is strictly connected with the parsimony principle, including many solved exercises that are classified according to the Donald Knuth ranking of difficulty, which essentially consists of a mix of warm-up exercises that lead to deeper research problems. A software simulator is also included.
The book presents a simpler unified notion of regularization, which is strictly connected with the parsimony principle, including many solved exercises that are classified according to the Donald Knuth ranking of difficulty, which essentially consists of a mix of warm-up exercises that lead to deeper research problems. A software simulator is also included.
Table of Contents
1. The Big Picture 2. Learning Principles 3. Linear-Threshold Machines 4. Kernel Machines 5. Deep Architectures 6. Learning from Constraints 7. Epilogue 8. Answers to selected exercises
Authors
Marco Gori Department of Information Engineering and Mathematics, University of Siena, Italy.Professor Gori's research interests are in the field of artificial intelligence, with emphasis on machine learning and game playing. He is a co-author of the book "Web Dragons: Inside the myths of search engines technologies,? Morgan Kauffman (Elsevier), 2007. He was the Chairman of the Italian Chapter of the IEEE Computational Intelligence Society, and the President of the Italian Association for Artificial Intelligence. He is in the list of top Italian scientists kept by VIAAcademy
(http://www.topitalianscientists.org/top_italian_scientists.aspx). Dr. Gori is a fellow of the IEEE, ECCAI, and IAPR.
Alessandro Betti Postdoctoral Researcher in the Department of Information Engineering and Mathematics (DIISM, University of Siena, Siena, Italy.
Alessandro Betti Ph.D. is a Postdoctoral Researcher in the Department of Information Engineering and Mathematics (DIISM) of the University of Siena (Siena, Italy). Dr. Betti's interests include analysis of algorithms, discrete mathematics, tree structures, and formulation of "learning laws? through least action like principles.
Stefano Melacci Senior Researcher (Tenure-Track Assistant Professor), Computer Science, Department of Information Engineering and Mathematics, University of Siena, Siena, Italy.
Stefano Melacci Ph.D. is a Senior Researcher (Tenure-Track Assistant Professor) in the area of Computer Science at the Department of Information Engineering and Mathematics, University of Siena (Siena, Italy). He has been the Research Manager of the Italian company QuestIT S.r.l. (Siena, Italy) and a Research Fellow of the Department of Information Engineering and Mathematics, University of Siena, where he received his PhD (2010), and the M.S. Degree (cum Laude). Since 2017 he has served as Associate Editor for the IEEE Transactions on Neural Networks and Learning Systems, and he is an active reviewer for several journals and international conferences.
His profile is strongly characterized by research activity in the fields of Machine Learning and, more generally, Artificial Intelligence. Recently, he has been working on new technologies for Machine Learning-based Conversational Systems and he studied and proposed Multi-Layer architectures (Deep Networks) for extracting information from static images and videos, using adaptive convolutional filters and principles from Information Theory. He previously worked in the context of Kernel Machines and Regularization Theory, under the unifying framework of Learning from Constraints that allows classic learning models to integrate symbolic knowledge representations. He proposed Manifold Regularization-based algorithms and Neural Networks that implement Similarly Measures, with applications to Computer Vision.
