- Artificial Intelligence: A Modern Approach by Stuart Russell and Peter Norvig Highly recommended
This is really the book to get on artificial intelligence. It’s extremely comprehensive — containing multiple semesters worth of AI, but every chapter is very well written, easy to understand, and (as a bonus) nicely typeset. I’ve seen this book on the shelves of many programmers who work with AI daily. But it’s suitable for anyone with an interest in the subject, starting out with the philosophical basis of AI but covering topics as advanced as information retrieval, neural networks and planning. The only complaint one might have is that all of the code listings in the book are actually given in pseudocode, but several implementations are available online at http://aima.cs.berkeley.edu/. Everyone interested in AI should strongly consider picking up a copy of this book for its breadth of coverage.
- Artificial Intelligence: Theory and Practice by Thomas Dean, James Allen, Yiannis Aloimonos Not recommended
This is a decent book on AI, but with a strong emphasis on LISP (in fact, all of the examples presented are in LISP). It goes into a great deal of depth in some topics — perhaps moreso than Artificial Intelligence: A Modern Approach, and includes more sample code. But the cost is that this book is far less readable. You should only pick up this book if you have a strong background in mathematics and don’t feel intimidated by complicated equations. As one student of Thomas Dean describes it, he’s the kind of guy who would call two rabbits alpha and beta when telling a story about his pets. I’d suggest browsing this book in a library or bookstore before picking it up.