Developing Games That Learn [Paperback]

Len Dorfman (Author), Narendra K. Ghosh (Author)

Book Description

Publication Date: March 1996

Now it's possible to improve a program's performance by getting it to learn from its mistakes. This practical book, and the companion program disk, present a high-level programming paradigm, "Objective Artificial Intelligence," that is used to devise algorithms which enable a program to learn. Presents three algorithms which can be modified to implement learning in many programs, and utilizes sample game programs to clearly demonstrate how these algorithms implement single trial learning. The program disk contains the complete C source code for the demonstration programs. For programming project managers, programmers, game designers, and game programmers.

Editorial Reviews

From the Back Cover

Now it's possible to improve a program's performance by getting it to learn from its mistakes. This practical book, and the companion program disk, present a high-level programming paradigm, "Objective Artificial Intelligence," that is used to devise algorithms which enable a program to learn.

Presents three algorithms which can be modified to implement learning in many programs, and utilizes sample game programs to clearly demonstrate how these algorithms implement single trial learning. The program disk contains the complete C source code for the demonstration programs. For programming project managers, programmers, game designers, and game programmers.

About the Author

Dorfman is CEO of MultiGrain Solutions.

Ghosh is a Senior Software Engineer at MultiGrain Solutions and is currently obtaining a computer science degree from Harvard University.

--This text refers to an out of print or unavailable edition of this title.

Product Details

• Paperback: 275 pages
• Publisher: Prentice Hall PTR; Pap/Dis edition (March 1996)
• Language: English
• ISBN-10: 0135696178
• ISBN-13: 978-0135696170
• Product Dimensions: 8.9 x 7 x 0.8 inches
• Shipping Weight: 3.2 pounds
• Average Customer Review: 1.5 out of 5 stars  See all reviews (2 customer reviews)
• Amazon Best Sellers Rank: #4,974,128 in Books (See Top 100 in Books)

More About the Author

Discover books, learn about writers, read author blogs, and more.

Customer Reviews

Most Helpful Customer Reviews

10 of 11 people found the following review helpful:

1.0 out of 5 stars Not worth the paper it's printed on, October 21, 1998

By A Customer

This review is from: Developing Games That Learn (Paperback)

This book claims to present a paradigm for implementing games that learn. In reality it's a book about a very specific implementation of a very specific game. The "paradigm" is limited to a few very general observations that you could have made yourself. 95% or so of this book is code, and this reviewer was less than impressed with the quality. Clearly, the code in this book was developed as a hobby project by a couple of non-professionals who decided to make it into a book. They also decided to ignore the vast body of existing work on artificial intelligence and make simplistic claims based on their own experience about what is and isn't feasible in a computer program. You should be able to find better books on this subject than this one.

4 of 4 people found the following review helpful:

2.0 out of 5 stars Not very professional, July 27, 2000

By 

Mihailo Despotovic (Silicon Valley, CA USA) - See all my reviews
(REAL NAME)   

This review is from: Developing Games That Learn (Paperback)

I agree with the first review (one star...). However, the book could be useful for someone who is starting to make his way in learning in strategic games. I would also mention that 3'5 floppy comes with the book (C code for DOS platform) so the fact that 80% of the book is a source listing is annoying. Begginers would find first couple of chapters interesting though.