Customer Reviews

Blondie24: Playing at the Edge of AI (The Morgan Kaufmann Series in Artificial Intelligence)

5 star:		(27)
4 star:		(6)
3 star:		(3)
2 star:		(0)
1 star:		(1)

 

 

An absorbing and enchanting tale of a personal quest for the deeper meaning of AI: the discovery of how intelligence itself arises. Fogel seizes the challenge by capturing the evolutionary process and shaping it to breed a checkers expert from an artificial neural net. Scientists, humanists, and artists will appreciate his inspiring wit and clarity of thought in narrating the growth of Blondie24, a synthetic sentience born inside a desktop PC.

The most enthralling account of what otherwise could have been a dry discussion of some arcane research. I have a couple of whimsical comments before I get on to the review proper.

David and Kumar played Blondie24, a computer program, against human opponents on a web site. They found that when they played using their own names, they were not invited to play many games. When they changed the name they used to that of a Star Wars(tm) character, they became more popular. However it was only when they started using 'Blondie24' as a name that everyone wanted to play them. This lesson was not lost on David when it came to naming his book...:)

In an early part of the book, David discusses the Turing test, which involves fooling a human observer into believing that a man, hidden from the observer, is actually a woman, based on responses to the observer's questions. Replace the man with a machine and repeat the experiment. David and Kumar seem to have run a little Turing test of their own. Humans at the checkers web site believed they were playing against another human, not a computer program. Indeed they were fooled into thinking they were playing a young woman, not a couple of frizz-haired mad scientists. (Please note I have no direct evidence to support my description of David and Kumar.)

An Artificial Neural Network/minimax program, Blondie24, learns to play checkers at expert level without being taught, without access to the vast human knowledge of the game. Significantly, Blondie24's game relies on n-ply position evaluation without recourse to an end-game database or opening book. The work of David and Kumar demonstrates that a solution to a problem is not always a necessary precursor to developing a computer program. Let the program find the solution on its own.

David's book is an entertaining and elucidating account of the development of Blondie24, a program that taught itself to play checkers. In it, he discusses why traditional AI research has been less than spectacularly successful and proposes a new direction: evolutionary computation. To prove the validity of this approach, he and his colleague Kumar develop a program that through the process of evolution through natural selection, learns to play checkers at expert level.

Blondie24 was put through its paces at a checkers web site, where it attained a level of 'expert'. Human players, who believed they were competing with a lady, blonde, 24, math major, were engaged in games over a period to determine how effectively the evolutionary computation approach had developed a competitive checkers player. This section of the book makes gripping reading, unlike this review.

I've had the opportunity to play Blondie24, which is now available as a PC game, and can attest to the strength of its game. There are four 'difficulty' levels, from novice to expert, so it suits all players. Not only does it play a great game, the presentation which revolves around film footage of your opponent, can be sometimes hilarious and sometimes infuriating. It is interesting to try and imagine what the program is doing while you are playing it.

I urge everybody with even a passing interest in the field of AI to purchase this book. Read it and then consider: is the genius David (and Kumar), who developed this program, or Blondie24, a program that taught itself to play checkers at a level much higher than its creators.

The serious AI researcher should also consider David's other books, also available at amazon.

Blondie24 is a fascinating and informative book that will be absolutely engrossing for anyone with an interest in artificial intelligence and computers. Although Master-level checkers programs have been around for a while, they have all used brute force to achieve their goals. The Blondie24 project represents the first serious attempt since Samuel's experiments in the 1950s to do something much more interesting and elegant: create a checkers program that can learn on its own. This book is easily accessible for the uninitiated, and I guarantee that you'll be swept along.

Gil Dodgen, author of the highly acclaimed computer program, World Championship Checkers

This is an engaging first-person account, oriented toward the general reader, of how a brilliant young researcher has expanded the envelope of artificial intelligence (AI). The author begins by describing how he has grappled with the fundamental issue of what AI should be. He, like an increasing number of workers who have parted ways with the traditional AI community, insists that intelligent systems exhibit the ability to adapt to changing circumstances in order to achieve goals. Astonishing as it may be to some, most AI systems have no adaptivity whatsoever, and a definition of intelligence as adaptive behavior is indeed a radical departure.

The rest of the story relates how the author (with his collaborator, Kumar Chellapilla) has applied his general principles of AI to the specific problem of having the computer teach itself to play checkers. The approach was to obtain a key component of a standard checkers program not by programming it with expert knowledge, but by allowing it to emerge in a process of simulated evolution. In this process, components were evaluated in play against one another, and the better ones survived and generated mutant offspring, while the others were discarded. I omit many details, but the essence is that this simple adaptive system gave rise to Blondie24, a checkers program that plays at the expert level. There are checkers programs that play much better than Blondie24, but none has ever played so well with so little input of human expertise.

In sum, the principles are potentially revolutionary, the practical results are extraordinary, and the reading is downright fun.

I was already interested in neural networks, and have read several other books on the subject, all for information on how to design and build neural networks to do various tasks, so I was primed to like this book, but I had no idea how compelling it would be. Fogel does give most of the details necessary to recreate Blondie, but some of the specifics for the later modifications are missing and can be found, I believe, in his and his partner's scientific articles referenced in the book.

The first half is quite good, giving very basic background information on neural networks and machine learning algorithms as well as a brief history of the two best known checkers programs, Samuels' machine learning checkers player and Chinook, the best ever player (so far).

It's the second half of the book, however, that kicks the story into overdrive. Fogel begins describing how he and his partner, Kumar Chellapilla, made their design decisions, evolved a neural net, then began playing it against humans over the internet. Even if you aren't all that interested in checkers (I am not) the games as they are described by Fogel become at least as interesting as any close sporting event, especially in those cases where the neural net set up traps for its opponents that neither Fogel nor the other player could foresee. There were times that I couldn't stop reading. I'd get off the bus to work, get in the office, and continue until I could pull myself away. The second half of this book is as much a page turner as the best novels I've read.

An absolute must read for anyone interested in where AI needs to go.

Anyone slightly interested in Evolutionary Computing and Neural Networks (Genetic Algorithms, Neuroevolution, etc.) is probably already familiar with David Fogel, or his father. This book is an exciting (dare I say adventurous!) read for anyone in the field of EC.

However, I think the biggest achievement here is the crossover value of this book. It appears to me that it should be readable and highly interesting to a broad spectrum of people, including those even only slightly interested in Artificial Intelligence, high technology, and chess, checkers, and other "deep" games (bridge?). This is mainly because the text is arranged in a unique way such that the muddy details of Blondie24 are easily skipped over (as they are in their own appendix). Dr. Fogel's clear, flowing, enjoyable writing style is also fabulous. Great for everyone!

I just finished this book yesterday. What a great read. The book very well written, I'd say it's often brilliant in places. Blondie24 is a computer program that taught itself to play checkers. As Fogel writes, the program used a combination of a simulation of evolution on his computer and models of how our own brains work to learn how to play. I find it amazing that a computer can teach itself to play checkers as well as human experts, and better than most people. What I really like about the book is that it's very readable. Anyone with a interest in science should be able to comprehend every idea in the book and it might just be able to change your opinion of what computers are really capable of. Not that this book hypes anything about AI, it doesn't. It plainly tells of the future of machines that can learn without needing humans to program in the answers. I find it all very fascinating and I think anyone with an interest in computers or science in general will too.

David has written an important and influential book. Not only is the discussion of Blondie24, the cute checker playing heroine of the book, a lively romp through the ins and outs of evolutionary programming but it lays the stage for David's more serious and far reaching discussion of what is right and wrong in our quest for a companion intelligence. It is this aspect of the book - David Fogel's keen insights into the meaning of machine intelligence and the ways in which we might grow and nurture our thinking machine of the very near future that I found the most arresting and thought provoking. In David's world view the AI community, by and large, has completely missed the boat - our current attempts to create intelligent machines by programming them to think is futile and, at the very least, misguided. Making short work of neural networks and chess playing programs, David uses the seeming simple but ultimately powerful example of Blondie24, a program that learns how to play checkers from pure trail and error - learning from experience in a process very similar to our own way of learning - to illustrate the power and range of possibilities in evolutionary programming.

This book is well written, easy (and , in fact, fun) to read and touches on topics that make it an important and necessary book for every intelligent reader in today's world of the internet and ever more powerful computers. I am impressed with David's well thought out observations on the role of evolutionary programming not only in the workings of Blondie24 but also as the core technology for the next generation of thinking machines. As the Chief Executive Officer of Natural Selection, Inc. David's views are more than academic musings, they are an important part of his company's business. Positioning himself as a new Darwin of evolutionary intelligence, this book - along with his previous book Evolutionary Computation - Toward a New Philosophy of Machine Intelligence -- places Dr. Fogel as the leading exponent of this rapidly emerging paradigm in machine intelligence.

While an absolutely must read for all of us in the Artificial Intelligence field, this is a book for any curious reader who is interested in how (and perhaps when) the thinking machines of the near future will come about. And, of course, Blondie24 is a smart, intelligent creature whose attempts to divine the workings of the checker board are fun to follow and satisfying to share.

This book tells you the story of a computer algorithm that, in the words of the author, "taught itself to play checkers". The second section, which contains the core of the book, is organized as an abridged diary that describes the steps taken to create this program and to test it in rated competition under the name of Blondie24 against human opponents at an online gaming site.

Since the book is aimed at a general audience, the first section introduces all the relevant knowledge required to fathom the rest of the book. This section introduces the purpose of artificial intelligence, the demystification of the all-famous computer Deep Blue that defeated chess champion Garry Kasparov, the basics of artificial neural networks and of the game of checkers, and a survey of previous attempts at producing computer programs to play checkers. The last chapter in this preparatory section states the two fundamental questions that the author (David B. Fogel) and his partner in this endeavor (Kumar Chellapilla) set out to answer when they started to build their own checkers program, which can be phrased as: (1) Can a computer program learn on its own the features important to play checkers at the level of a human expert? (2) Can this learning be achieved by just playing games against itself and receiving feedback only after a series of games without even knowing which games were won or lost but only how many?

Their approach consisted of using an idea borrowed from "mother nature" that only until recently has started to be embraced by the scientific community in the field of artificial intelligence. This idea is evolution. By combining random mutation with selection over a "population" of checkers-playing artificial neural networks that played against each other they obtained after 250 generations a program that was able to reach the expert-level rating and that even scored a few victories against human players rated at the master-level.

Most technical details are left out in order to make the text accessible to a wider audience. However, in the spirit of being a scientific document, there are references to all relevant scientific papers in case you want to do further research. The writing style is both engaging and easy to follow. In addition to the main text of the book, there is a wealth of notes in a special section at the end of the book which the author uses to expand on specific topics that might be of interest to the reader. It is for this separate notes section that you might benefit from using two bookmarks, instead of just one, while you read the book. There is also an interesting section in which the author addresses a series of objections that have been raised against the ideas he discusses in this book.

In my opinion, the only weakness of the book is that it spends too much ink in telling you about the moves that Blondie24 (and its siblings) made in some specific games. This will be of interest to you only if you are keen on checkers.

If you have an academic training equivalent to a B.S. in Computer Science you will have the additional benefit of ending up with a clear picture of how to reproduce the ideas used to create Blondie24, although no computer code is offered. The idea they use is so simple and yet so powerful that you'll be temped to jump into the bandwagon of evolutionary computation after reading this book.

David has produced an important, and inspirational book. AI has received a bad press in the past. Many claims were hyped and the promise never delivered. David has shown that the claims of AI are sometimes true. The book first introduces the history of computer checkers, based on hand-coded solutions developed by expert players. The second half of the book describes the process of developing the artificial neural network player, Blondie24, through a process of evolution. The player that evolved had no human expertise to rely on and learned to play the game through trial and error only. This enthralling book demonstrates that evolution is a powerful force in the search for intelligence. You soon start to believe once you have lost a game to Blondie24.