Description
The second edition of Machine Learning: A Constraint-Based Approach provides a fresh look at the fundamental models and algorithms of machine learning, based on the unifying concept of “learning from environmental constraints.” The authors attempt to provide a unified framework for understanding machine learning, focusing on current topics of interest such as neural networks and kernel machines. In this approach, symbolic knowledge bases are considered as sets of constraints, and a path is outlined for deep integration of machine learning with multivalued logics, such as fuzzy systems.
Special attention is paid to the topic of deep learning in this edition, as it fits well with the constraint-based approach followed in the book. The book introduces a simpler and more unified concept of regularization, which is closely related to the Parsimony Principle. This unified structure makes it easier for students and professionals at the master’s level and above to understand complex concepts.
Book Features
- Presenting fundamental machine learning concepts (such as neural networks and kernel machines) in an integrated, constraint-based manner.
- Deep coverage of unsupervised and semi-supervised learning, with new content in fast-growing areas such as deep learning.
- Includes hundreds of solved examples and exercises, categorized according to Donald Knute’s difficulty rating.
- Along with a software simulator for nuclear machines and learning from constraints to gain experimental skills.
- Introducing the integrated concept of regulation related to the principle of thrift.
Book specifications
- Publisher: Elsevier
- Instructor: Marco Gori
- Number of pages: 549
- Number of chapters: 8
- Format: PDF
Headlines
- The big picture
- Learning principles
- Linear threshold machines
- Kernel machines
- Deep architectures
- Learning with constraints
- Epilogue
- Answers to exercises
Pictures
User Guide
Extract the file and run it with the appropriate software.
Download link
Download Elsevier – Machine Learning: A Constraint-Based Approach, 2nd Edition 2023
File(s) password: www.downloadly.ir
File size
4 MB
