|
The Emotion Machine: Commonsense Thinking, Artificial Intelligence, and the Future of the Human Mind | 
| Author: Marvin Minsky
Publisher: Simon & Schuster
Category: Book
List Price: $26.00
Buy New: $5.95
You Save: $20.05 (77%)
New (9) Used (10) from $5.10
Avg. Customer Rating:  20 reviews
Sales Rank: 10134
Format: Bargain Price
Media: Hardcover
Number Of Items: 1
Pages: 400
Shipping Weight (lbs): 1.3
Dimensions (in): 9.5 x 6.3 x 1.4
Dewey Decimal Number: 153
ASIN: B000WPPYGS
Publication Date: November 7, 2006
Availability: Usually ships in 1-2 business days
Shipping: Expedited shipping available
Shipping: International shipping available
Condition: Brand New with small remainder mark. Never read/open. Ship daily @8:30am w/ delivery confirmation.
|
| Also Available In:
|
| Similar Items:
|
| Customer Reviews: Read 15 more reviews...
 Worth the read. July 6, 2008
Minsky presents interesting new ideas on understanding ourselves. It makes sense that the mind, like the body, may seem simple on the outside but is amazingly complex on the inside.
 Very Interesting subject. Good discussion points. May 1, 2008
Like all books on human intelligence also this one is obviously speculation. But it is very helpful speculation if you are interested in the subject. Not all is simple reading especially if you have not been exposed to it before. My own opinion was well advanced by reading this book and I too have agreements and disagreements. Minsky properly suggests that looking at the mind in a too fragmented way will stop us from understanding it. I too see the mind as a complex whole of concepts, processes, and most of all interactive neural resonances. Minsky on the other hand then goes on to fragment the mind's functioning into 20 different capabilities and offers proposals how they work, but enough how they might interact. His perspective is still too fragmente, but then he suggests that uilding those functions into software might create an artifical intelligence.
My own understanding was the most enhanced by his description of how motion and sensation form an expectation loop, meaning that each action has a connected sensation response that leads to another action and so on. This motion/sensation mechanism is built into neural nets as are all other functions of the brain. It allows us to act without abstract thought.
The connection of motion/sensation with abstract thought (often referred to as reason or logical thinking) and emotions requries the contemplation of the human limbic system and neurotransmitters. There I miss quite a few points to be made that Minsky missed.
I also disagree with his expectations into AI. It is this complex resonance between action/sensation, abstract thought and emotion/feelings that creates not only our human experience, but also our intuitive capability. Therefore that will not be emulated by an artificial intelligence mechanism as it lacks our human biological experience that shapes us so strongly. I suggest Steven Pinker's The Blank Slate: The Modern Denial of Human Nature for a better understanding of how we become humans.
My own thoughts of what an AI computer might turn out to be like and how cruel and distorted too rational human beings can be you will find in my own novel Deity. Cruelty is not emotional it is rational.
Overall 'The Emotion Machine' is a great book on the subject and well worth reading if you want to expand your own thinking. The times when a single book was supposed to contain all the dogmatic explanations for our life are gone.
 I am not its target audience March 12, 2008
I put this book on the shelf last night about 2/3 done, it will not be finished. I have a back ground in controls engineering and physics simulation, and much of what the author talks about is very similar to these subjects (feedback-control, gain schedualling, non-linear systems, constrained optimization, multi agent interactions etc) yet there is absolutely no use of this body of knowledge. Also the 'style' used in the presented flow diagrams is pain full.
From the word 'machine' in the title I had expected a book with at least a hint of pseudo code or math but got none. I think if you have 'cut metal' on anything related to making a computer do anything that a human can do you will not enjoy this book.
Disappointing and too "Textbook-like' February 18, 2008
Several other reviews do a good job at describing the content of the book. In contrast I'd like to focus on my reactions to it.
I have read about conciousness before -- including the works of Hofstadler and Penrose ("Emperor's New Mind"). I expected "The Emotion Machine" to be a similar work, exploring the issues of conciousness and the mind in a way that would stimutae further independent thought and ideas about an critical scientific/philosphical issue that mankind and science has been grappling with for ages.
For further personal background, I also have a doctorate in the hard sciences, so I don't think that mist works need to be "dumbed down" for me to understand. Though I do admit there are many smarter and more dedicated people in the world to these issues than myself -- where I'm closer to the "educated layman."
Unfortunately I found this book to be far too textbook-like, much as it was intended not to stand alone but to suppliment lectures and discussions on the topic. I kept getting the "Is this going to be on the test?" feeling and found that this focus did not allow me to shift my focus to the big issues surrounding the mind.
Perhaps the book was intended to be for a different audience than myself, certainly there are other reviewers that found it valuable. For myself however, though given the jacket wording, back cover comments and site summary indicated otherwise, I felt that the book was not valuable to me.
If you're interested in the mind and aren't scared away by some mathematics and scientific method, go instead to the works of Penrose and Hofstadler. This book can be given a pass.
Society of Mind II February 7, 2008
A good book with interesting ideas. However, there was a fair ammount of rehash from his other book (Society of Mind), there was alot of conjecture with not much expereimentally backed up theory, and of course no sourcecode!
|
|
|
Powered by Associate-O-Matic
| |
