Customer Reviews

Building Problem Solvers - Listings

5 star:		(2)
4 star:		(0)
3 star:		(0)
2 star:		(0)
1 star:		(0)

 

 

Building Problem Solvers is a very hands-on introduction to AI systems in general, and truth-maintenance systems in particular. All the systems discussed come fully implemented. In this regard, the book can be viewed as an accessible and detailed discussion of the code. For this reason, the best way to read any chapter is to scan through it first, then, carefully go through it again with the source code in sight.

The authors were PhD students of Gerry Sussman at MIT. Thus, this book is a great way to learn about the classic AI systems and techniques devised and refined at the MIT AI lab.

If you get serious about the book, you'll want to try out a few of the exercises. I found that the exercises are invariably insightful, though I wish some included implementation hints, because elegant solutions are often far from obvious.

In short, I highly recommend this book if you're looking to build some problem solver using proven AI techniques.

This book is very rewarding to study and put into practice. I had a lot of fun immersing myself in the concepts and systems developed in this book. Using the techniques of this book as a base, we've implemented BioHacker, a debugger for metabolic networks.

This book takes you on a journey through the techniques that have been used to build problem solvers, ranging from classical search techniques to Assumptive Truth Maintenance. It is very easy to read, despite the high level of technical detail. The LISP code that accompanies the book is well documented, easy to understand, and it works. It is a must have for anyone who truly wants to understand problem solving techniques.