Artificial Intelligence Books

I had high hopes for this book as part of my Artificial Intelligence course. I've been unhappy with the edited volumes where I would find a small subset of papers that I wanted to use in a course. Copeland's treatment of the intersection between Philosophy of Mind and Artificial Intelligence seemed to be exactly what I was looking for in a text. Unfortunately, the AI content is extremely dated, making it nearly useless to my computer science majors. (If I were teaching in a Philosophy department that wouldn't matter as much.) The first half of the book is great as a historical perspective, but I'll be going back to the edited volumes next time I teach AI and want to cover the Philosophy of Mind questions.

This is a fascinating and lively book, which is almost incredible give that it is an introductory philosophy textbook. Copeland manages to write with both personality and balance. The combination of his style (which is clear and witty without being facetious)and the intrinsic interest of the subject of artificial intelligence had me hooked. I read it like a novel, never wanting to put it down. Copeland assumes no prior knowledge of computer science, psychology, or philosophy, so the book should be accesible to any intelligent reader, although a few parts can be hard going. Beginners are likely to struggle with the sections on the CYC project (in chapter 5) and the Church-Turing thesis (in chapter 10), but slow and careful reading should do the trick. Copeland does explain eveything you need to know in order to understand what he's saying, but some of his explanations are gentler than others.

Otherwise my only complaint is that Copeland raises some interesting questions without exploring them very far. His view on the prospect for artificial intelligence is that, given the purposes for which we use such concepts as thinking, it is quite possible that there will come a day when the only reasonable course is to say that machines can think. In other words, he thinks that computers cannot now think, but that one day they (or their descendents) might become sophisticated enough that we ought to change our use of the word 'think' so that it applies to machines as well as humans. But he says very little about the purposes of concepts like thinking. In particular, he ignores the idea that rationality (surely a related concept) has great moral significance of a kind that might well make some people highly reluctant to say of any machine that it really thinks. Since this is an introductory book I don't hold this against Copeland, but it would be nice if he would say something about this in the next edition, which I believe is due out soon.

I'm looking forward to it.

The book is good, covers a lot and is very clear.

The downside: there are some small errors and mistakes.For example, the authors define gamma: SxAxE -> 2^S as the transition function, where S is the state space, A is the set of actions, E is the set of events. Later they say that if there are no events to be considered from the outside world, then you could use E={} (empty set) -- Assumption A3, page 10. Although this is intuitively OK, it is mathematically flawed, because the cartesian product of anything with {} is {}.
Planning with MDPs and specially with POMDPs deserves more attention. In particular, the very short commentary on planning with POMDPs mentions that it is not possible to solve big POMDPs. This is not true anymore; there are very good heuristics for POMDP solving currently.
I think more theorems could have been presented and proved, and some advanced sections could be added to each chapter (some authors include a section with a star, for example)
I also don't like the way pseudo-code is presented, but that is a matter of taste.

It would also be nice if the examples in chapter 2 were fully specified. That helps a lot to understand how problems are represented.

On the good side, there are LOTS of examples for each definition, and there are exercises at the end of each chapter (more exercises would be nice, actually). I also like the discussion and historical remarks at the end of chapters.

This is certainly a very good book. Anyone interested in planning ought to have it (and people interested in AI will certainly benefit from it).

Automated Planning is a good book for those who get started out in the field of search and planning. It's a good overview of the topics that abound within the planning community.

The only downside of the book is its dealing with important topics like planning graphs and markov description process is cursory, and more detail would have been nice.

Can one take any theorem or conjecture in mathematics and translate it into the language of OTTER in order to either prove the theorem or verify or disprove the conjecture? Unfortunately, the answer is no, as the translation of most of the statements of mathematics into the clause form required by OTTER will take a lot more research to bring it into fruition. The success of such research would be signal a major breakthrough in automated reasoning or automated theorem proving, as one could engage machines in the generation of new mathematical results and the settling of outstanding conjectures.

A variant of OTTER, called EQP, resolved the "Robbins conjecture" in 1997, via the efforst of the mathematician/computer scientist William McCune. First studied in 1933 by Herbert Robbins, the conjecture asserts that every Robbins algebra is Boolean. The proof took 8 days to complete and made international headlines. The success of this proof motivated many to look more deeply into OTTER, and since it is in the public domain, anyone curious about it can obtain it and use it. This book gives a comprehensive and very understandable overview of OTTER, and can be read by anyone with a background in mathematical logic. Some knowledge of logic and functional programming will help too. In relation to OTTER, one goal worth pursuing is to find out to what extent various fields of mathematics can be translated into the (clausal) language of OTTER. Point-set, geometric, and algebraic topology come to mind, as well as algebraic geometry in the guise of schemes and functors of points. Some automated proofs have been found in topology and algebraic geometry. It remains to be seen whether most, if not all the concepts in these fields can be expressed in a language that will enable automated proofs to be given.

OTTER is an ancronym for Organized Techniques for Theorem Proving and Effective Research", and after a forward to the book by Larry Wos, the author gives an introduction to the language in chapter 1. His concern is with applied logic and not theoretical developments, so the presentation is informal, and therefore useful to those who are anxious to learn OTTER and apply it. Theoretical developments are not completely ignored though, and throughout the book one can see to what extent expressions can be regarded as clauses and then translated into the language of OTTER.

This book fills a major void in the existing literature on OTTER,
the automated reasoning program developed at Argonne Research by William McCune. The book seems unusual in that, on one hand, it provides numerous examples of OTTER input/output files, useful tips on operating OTTER, and a plethora of exercises which, if carried out, will lead to a rapid understanding of the program and how to use it. But on the other hand, the book also works as a formidable introduction to automated reasoning. Starting from basic concepts such as inference rules and unification, and working up to more advanced topics in equational reasoning, one can gain a fairly good introduction to the theory of automated reasoning. My only complaint involves the lack of a good appendix or glossary which lists and summarizes the numerous commands that can be fed to OTTER. Furthermore, many of them are not even indexed, which makes referencing them somewhat tedious. Other than this, I highly recommend the book. However, I would encourage the novice to first study a more user friendly logic programming environment, such as swi prolog, before attacking OTTER. For having some experience with prolog programming will allow the reader to compare and contrast the two automated-reasoning methodologies. As for OTTER itself, I consider it more useful than a prolog interpreter since it allows for the use of
strategies for finding the desired proof or computation. On the other hand, it is a living embodiment of the fact that there is much progress that still needs to made in developing useful and powerful tools for automating logic and mathematics.

This book is a direct descendant of the Hampdenshire Wonder, although in this case the enhanced intelligence is controlled and deliberate.

When a very young James Holden's parents are killed, he is left in a different situation to most children in this situation. His clever scientist olds had invented an education machine that leaves him at five with the intellectual development and education of someone in high school, and beyond.

Much as Tim does in Children of the Atom he realises he can support himself by writing, and makes a living doing so when eight years old. This is a profession where you do not need to be seen.

Eventually he needs an adult front for economic reasons, and he approaches his landlord.

However, he can't stay hidden forever, no matter how clever, and unscrupulous types, as well as the government and the judiciary with ambition decide to put his invention to use, while shielding him from harassment, spies, and other such annoyances.

Terrifically suspenseful and readable tale of a small boy artificially made super-intelligent by scientists. He escapes his keepers and uses his superior brain to evade authorities and get along in the adult controlled world. If you liked James Patterson's "When the Wind Blows" and sequels, you'll like this original, tightly written suspense novel even better.

Having been involved in the technology business for way too many years, I found Newquist's book to be a pretty accurate account of what went on in the AI world. It's not a book about technology or programming, so if you're looking for technical data, try a textbook. But if you want to know why artifical intelligence died a slow and ugly death, you'll have all your questions answered here.

The book, as its subtitle suggests, is about "genius, ego, and greed"--the personalites involved in AI. It's not about the importance of neural networks vs the relevance of expert systems. As for the "discot" review that says to take some of the information with a grain of salt, Newquist includes nearly 15 pages of footnotes to back up his research. That should be good enough for most readers.

All in all, I found this book to be an insightful observation and reflection on what AI could have been. I'd recommend it over books by AI participants like Raymond Kurzweil, who obviously have personal motivations to keep selling AI snake oil in their self-promoting books.

I found this book both interesting and limited. Having encountered AI technologies and personalities at various points in my career, it was interesting to get from the author a more comprehensive view of what was going at the personal and organizational levels during the period covered by the book. The author's view of these matters was culled primarily from his experience as editor of a newsletter called AI Trends during that period. I visualize the author writing this book by pulling from stacks of old newsletters, article clippings, corporate brochures, and notes from interviews and discussions he had as a journalist on the beat. I see this book's value mainly in that it summarizes a lot of information about AI people and organizations in one place, organizing it into thematic chapters.
The author inserts his own perspective throughout the book, with mixed results. He is attracted to the dirt, the scandal, the quirky personality, and this leads to some interesting reading, interesting in the way you might listen to the town gossip, in spite of yourself. I had to take his gossip with a grain of salt, because some of it was based on questionable interpretations of the author, but enough was substantiated to be interesting. For example, the rise and fall of AI companies is an interesting story that parallels that the recent dot com cycle, and the AI era has lessons to teach us about the business and management of technology. However the author's bias toward airing dirty laundry sometimes comes across as a sneering attitude, or at least over-dramatization, and some of the ugly pictures he paints seem ugly because of his paint, not the events he reports. For example, he presumes to classify management talent as "A-teamers" (capable) or "B-teamers" (less capable), then identifies hiring B-teamers as evidence of poor management in some companies.
The author clearly does not have a deep understanding of AI technology, and this limits his ability to achieve two things he tries to do in the book: (1) explain AI in laymen's terms, and (2) interpret the technical significance, shortfalls, and potential of AI technology. He is on target some of the time, and sometimes misleading, or even wrong. For example, as the author correctly points out, the publication of the book Perceptrons by Minsky and Papert was an intriguing chapter in AI, since it effectively shut off research in neural networks for a long time. However, his discussion of the essence of Perceptron's criticism of neural networks is misleading: he says it was that neural networks cannot ".. learn new things from past experience..", when actually the main criticism was that certain kinds of problems can never be solved by neural networks. His discussion of how researchers eventually countered Perceptron's arguments is also misleading: he cites Hopfields's showing that recurrent neural networks can do things the brain does (an important contribution), when the more relevant direct answer to the Perceptron dilema was the development of good training algorithms for multi-layer nets that could solve the "impossible" problems.
I enjoyed reading this rather long (488 pages) book. It moved along quickly, and it was interesting to find answers to a lot of "whatever happened to ..... ? " questions. The AI era covered by this book was filled with fascinating stories and people. I would have preferred a more penetrating and knowing analysis of AI technology itself, that would make it easier to separate good ideas from business blunders, circumstances from fundamental flaws. And I wanted to have less of a feeling that the author was just guessing at some of his insider insights.

Freedman's book is a useful introduction to artificial
intelligence suitable for the lay reader. More technical
introductions exist in the form of the many good textbooks
on AI.
Brainmaker is a fun read but I do have a few criticisms.
Freedman distinguishes "old time" "good old fashioned AI"
g.o.f.a.i. from what he dubs "nature AI." I believe that he
has simply cobbled together some ideas and that his "nature
AI" does not exist as a coherent project.
Freedman seems to think that gofai was not modeled after
nature. I do not agree. Newell studied how people reasoned,
Boole was building a logic of how people think, and Rosenblatt
had real neural nets in mind. It is also inaccurate to call
gofai a failure. Sure there are lots of things that people
can do that computers can't. But there is also a long and
growing list of what computers can do and people can not. AIs
are good a modus tollens, humans are not. AIs are good at
long chains of reasoning and with negated terms, humans are not.
Computers can handle spaces having many dimensions, humans find
it hard to handle 3. Computers are good at probability and
math, people are not. And the list goes on and on.
Freedman's "nature AI" does not exist as a real AI subfield.
He has simply grouped together a number of new ideas, some good,
some bad. He also spends too much time on biology. If
there is any evidence that intelligence requires "wetwear" such
evidence is not presented in the book.

If you've ever read 'I, Robot' by Issac Asimov (or anything like it), this book should be required reading for you. It provides a fascinating and candid view into the history, research, debates and a detailed summary of the current progress in the controversial field of A.I. (artificial intelligence). I thoroughly enjoyed this book, and have shared it's title with many who have also enjoyed it. As an author and sci-fi fanatic, I found it's forthright, non-technical manner both refreshing and easy to read (finished it in one night), and the information provided to be excellent food for thought. Not too technical, this book can easily be understood by just about anyone, and SHOULD be read by anyone who has ever enjoyed reading anything involving robots or computers of any kind (even the average Star Wars fan will like this one). 'Brainmakers' is a must-read which has earned a permanent place in my personal library (which I might add is fairly extensive and somewhat exclusive). If you can find a copy, it's worth whatever price you have to pay for it.

Really neat book. I have wanted to tinker with an underwater robot for some time. Indeed, who hasn't wanted to build a submarine or something like it since they saw their first submarine war movie.

This book presents the fundamental design challenges in very clear English. There are 2 vehicle designs presented with plenty of diagrams and hints to avoid common pitfalls. The techniques for waterproofing the motors and control system are very simple and are commonly available in any hardware store or even already in the household. Great book to get one started in underwater robotics. And to finally fulfill that childhood dream.

This is indeed a great book, but guys... $192 ! You can buy this elsewhere on the net for around $30 including delivery. Don't give in to gougers, it just isn't that rare a book !

provide a good and easy to understand introduction to the subject

This book gives a good introduction to the mathematical modeling of cognition and does so with a level of mathematics that is very accessible to a typical graduate student in computer science or psychology. The book could have been written using tools from measure theory but luckily it was not for a book at an introductory level. The concept of probably approximately correct is introduced early on in the third chapter of the book with efficient learning given later on in Chapter 5. Chapter 7, the best chapter of the book, discusses the idea of VC dimension, which has had many applications, such as network stability and optimization. VC dimension plays the pre-dominant theme in the rest of the book, with the book ending with an application to neural networks. There are short problem sets at the end of the chapters, and these are useful for more understanding of the concepts in the book. A very interesting book and worth the price.

I found the projects interesting and the introduction to Not Quite C NQC excellent. The author demonstrates how to connect the touch sensors and light sensors physically and then add programming code to control the logic that produces power from the central processor to a motor that powers a gear. The basics are simple to understand and the project increase in sophistication and complexity.

The book explains explains different types of gear ratios, differential gear combinations, power and speed gear ratios, pulleys, power connections, stop latch, levels, frames, sound, and how to simulate an actuator like a grabber.

The robotic behavior can be either conditional or remote controlled. I'm a programmer and this book helped me break into the world of robot programming, signal programming, multitask abstraction, and signal processing without having construct the hardware.

Lego Mindstorm is a much easier and faster way to build simple robots verses trying to construct all the hardware on your own. Each chapter has a flow chart of tasks and functionality that help explain the logic controlling the robot. I found this book a delight to read and understand.

This is a first impression, and may be upgraded later. The textual content, including appendices, is good. Unfortunately, the illustrations have inadequate contrast, and are difficult to interpret even under very bright light. This is due to extensive and unnecessary use of gray scale, both in assembly drawings and flow charts(!). While it is quite possible that Apress printed the grayscale much darker than Mr. Baum intended, I believe the decision to use gray scale at all was flawed. On the other hand, the use of isometric ("3D") projections is entirely appropriate. If reprinted with isometric line drawings (no gray scale) I'd probably rate it 5 stars (especially if I could trade in my gray scale version).

What do musicians, artists, linguists, Buddhist monks, and computers have in common? This book takles this broad scope with some very interesting revelations. The author has a Ph.D. in computer science and an undergraduate degree in eastern & western philosophy. Thus, the title and the Buddhist angle.

This books reads easily and is very entertaining. Coming from an engineering background, I appreciated the author's structured writing style. That is, he does not meander or get flowery with his words. He states his facts, makes his points, and moves on. The reader does not get overwelmed with too much detail or historical data, but an extensive bibliography is available for the curious. The plenitude of charts and illustrations is helpful and at times a necessity.

In the beginning of the book the author keeps each subject separate: one chapter dedicated to linguistics, another chapter to abstract art, etc. Slowly he begins to reveal how all these areas mesh, which left me anticipating a climatic revelation that would tie it all together. However, I found the conclusion to be somewhat anti-climatic involving the future of virtual reality and the author's own eastern religious beliefs.

If you are a musician, artist, linquists, or work with computers and you have ever wondered why you think the way you think, definitely read this book. It will open your eyes and broaden your horizons immensly. Caution: if you are looking for "how to" information, technical details, or references to the latest/greatest equipment, it's not what this book is about.

This beautifully written book offers interesting sections into the history of algorithmic thinking in the arts, and builds a concept of the computer use in art thereon, embedded in a context of indian mythology. It is the best book combining music and visual arts in this respect that I have read so far and its concepts will certainly and hopefully be quite influential on the producers of new media work. Musicians might find the chapter on serial music a little superficial, as visual artists might perceive the one on Kandinsky's work, and I am not so sure whether I agree with the author's personal "unified theory" presented in the last chapters of the book, but the strength lies in the combination, and if you are looking for a general introduction, it is a stimulating overview that serves as a great starting point for further studies. In comparison to Hofstaedters "Goedel, Escher, Bach" this one feels more relaxed and undogmatic and stays with topics of abstract language and generative grammars, instead of trying to hammer a cynical anti-spiritual pseudo-religion into your brain.