I think that AI might be a precious tool in the developer's toolbox, and I'd like to know more about this field hoping that it will make my life easier as a developer.
Can you recommend some classics, books, links, essays, authors, reading, and bloggers? What is the Code Complete of the AI field ?
"Programming Game AI by Example" is a very good book for game AI containing both theory and practical examples and algorithms.
Check it out: http://www.wordware.com/files/ai/
For some online resources you can check out: http://www.aaai.org/AITopics/pmwiki/pmwiki.php/AITopics/StudentResources
"The Art of Computer Programming" by Donald Knuth.
There are lots of AI and other algorithms and this book will "turn your brain" in correct direction.
Goldberg, David E (1989), Genetic Algorithms in Search, Optimization and Machine Learning
Machine Learning by Tom Mitchell is a bit on the older side, but almost everything in there is still just as applicable now as it was in the 80's. AI is the sort of field where a good, broad understanding of core concepts is needed in order to be able to work with the more advanced/cutting edge things.
The book also goes over a bunch of good experimental methodology stuff as it relates to AI so you can test how well your algorithms work.
Artificial Intelligence: A Modern Approach by Stuart Russell and Peter Norvig
Programming Collective Intelligence by Toby Segaran is a modern, hands-on guide with real world example applications. The examples are in Python, but it's still worthwhile even if you program in other languages.
Machine Learning by Thomas Mitchell
This book is a classisc and a must for anyone who want to learn about machine learning.
Artificial Intelligence : The Modern Approach by Stuart Russell & Peter Norvig
This is a book that touches on a lot of topic on AI and machine learning. This is probably the closest to "Code Complete" of AI. Peter Norvig, one of the authoer, is now working at Google.
There are a lot of sub-fields of AI including Machine learning, Data mining, Computer vision, Pattern Recognition, Natural Language Processing, Genetic Programming and etc so I probably could not list all the books here but if I have to pick some of the most influential books, I would pick the two above.
http://en.wikipedia.org/wiki/Prolog and also for the intro Kowalski, R. A.. The early years of logic programming.
Stanford has posted several courses on-line (lectures + materials) including one on Machine learning.
Search google with "stanford engineering course machine learning", the first URL is the machine learning course. The lectures are available from there or even on itunes.
Or from here : http://academicearth.org/courses/machine-learning
(sorry for no url, unable to add hyperlinks as a new user)
Elements of Statistical Learning by Hastie, Tibshirani, and Friedman surveys just about every Machine Learning method worth knowing while also developing some unifying theory.
It doesn't directly address computation but it has lots of exercises that you could use to figure it out.
I'm quite suprised not to find Pattern Recognition and Machine Learning from C.M. Bishop. But yes, definitely in my top 3 with Thomas Mitchell and Stuart Russell & Peter Norvig.
This was my answer to a similar question:
AAAI is an excellent site that has all the important resources for students including discussions, links, videos and books. Highly recommended.
An awesome survey of all the important sections i.e learning, planning, problem solving techniques, including some philosophical implications are discussed in the AI:Modern Approach. The writing is very lucid ,the chapters are self-contained , you can select what you like and start hacking immediately. The site also has links to courses that used this book so you can get a lot of cool project ideas, assignments etc.
Peter Norvig's PAIP is another good book. I have not read this but many of my friends say that the projects in the book are top notch and some even claim that this is the best programming book ever. It should be fun to implement Prolog. However you should know Lisp. There was an attempt to port it into Ruby by Olan Bini.
The open directory project also has good handpicked resources ,books ,links to softwares and algorithm implementations that you can use in your project.
I have a wishlist in Amazon if you are interested.
Or you could see lists like these at Amazon
Machine Learning at Stanford.. Awesome Professor.
Stuffs at Internet Archive. Old historical discussions.
The Machine Learning subreddit has interesting links for all levels.
What Computers Still Can't Do: A Critique of Artificial Reasoning
By Professor Hubert Dreyfus.
It is a thought provoking philosophical assessment of AI. It approaches the problem in much more fundamental, phenomenological terms and outlines much of the why and how computer AI can't surpass human experience and understanding. A useful counter balance to the hubris of AI research. While not very technical at the very least it helps to understand the domain of problems that are only really solved by humans and why exactly it is that only humans are able to solve certain problems.
You might want to check out Heaton Research. We host the Encog project, and also have quite a few articles to read on neural network programming. The site focuses primarily on C# and Java. We also host the Encog project which is a neural network framework for C# and Java.
One of the best book I have read on AI that should be read in first is :
Society of Mind - Marvin Minsky
It gives to you a perfect introduction to ideas and concepts of artificial intelligence.
As already said, Mitchell and Russell&Norvig are the 'best first' :P
I'd like also to highlight Ivan Bratko - Prolog Programming For Artificial Intelligence, a must for Prolog programmers and a supernice compendium for AI students (the spectrum of AI topics is complete and coherent :P)