Artificial Intelligence: Theory and Practice

Author: Thomas Dean
Publisher:
Publish Date: 2004-06-01
Features: This book emphasizes the importance of expression in core chapters on logic, search, and learning. Compared to most introductory textbooks, it introduces more formal discussions of artificial intelligence, evident in its focus on syntax and semantic logic as well as its coverage of the computational complexity of AI algorithms. New results in computational learning theory are included as supplementary material to illustrate various techniques ranging from decision trees to neural networks. The book provides examples of robots and software automation (software robots) as teaching cases in real-world AI, along with a wide range of other examples to highlight the potential and diversity of AI applications. Chapters on natural language processing, planning, uncertainty, and vision integrate existing methods, summarize challenging areas, and outline technical prospects for further research. The book does not aim to cover all aspects of AI technology, merely mentioning topics like qualitative reasoning in physical systems and analogy reasoning in passing, while giving more attention to other subjects than traditional textbooks. It delves into learning, planning, and probabilistic reasoning to reflect their growing importance in the field. The chapter on vision (Chapter 9) addresses topics crucial for understanding perception in intelligent systems and the development of artificial systems that interact with the world in practical and engaging ways. This is a textbook meticulously compiled by three senior AI experts, explaining both the fundamental theories and practical applications of AI. It introduces new research findings and discusses practical issues in system implementation, addressing expression and computational challenges in machine intelligence system development. The authors explore traditional symbolic reasoning techniques for solving learning, planning, and uncertainty problems, such as deductive reasoning and decision trees, while also introducing new technologies like neural networks and probabilistic reasoning. Important algorithms in each chapter are accompanied by their LISP source code implementations for reference during experimentation. Additionally, the book provides numerous examples of AI application systems. It is suitable as a textbook for AI courses in computer science, control engineering, mechatronics, mathematics, and related disciplines at universities and colleges, as well as a reference for researchers and engineering professionals engaged in AI research and applications.