Artificial Intelligence: A Modern Approach by Rusell and Norvig is more or less the standard AI textbook and the book I'd suggest to get an overview of AI and its different methodologies. Mind you, it's over 1000 pages, but a very interesting read.
I reject that. Russell and Norvig really turned me off to AI, because they mainly come from a perspective "good old-fashioned AI" (GOFAI). This mindset is more about discrete symbols and logics, and reject uncertainty, probability, and fuzziness.
I was turned off by the entire field until I began learning about statistical, empirical, and data-driven approaches.
I heartily endorse Bishop (2006). It's a much more modern treatment.
Make sure you know basic calculus, and understand derivatives. More math is always better, of course.
"I have deeply regretted that I did not proceed far enough at least to
understand something of the great leading principles of mathematics,
for men thus endowed seem to have an extra sense." --- Charles Darwin
Here are some classics in the field. I'll let you google them yourselves.
LeCun et al, 1998: Gradient-based learning applied to handwriting recognition. A deep convolutional net that can read handwriting, and was deployed nationally . Yann LeCun tells me that a patent lawyer in California reimplemented in his free time as a hobby, so it can't be that hard.
LeCun et al, 1998: Efficient BackProp. Tricks and implementation details that are not discussed often.
Btw, as I understand it LeCun was offered a position to be head of Google research. He declined, and Corinna Cortes took the job instead. Regardless, if you googled Yann for a while, Google ads would try to entice you to work at Google.
There is a recent trend in neural networks towards something called Deep Learning. This deep neural networks more closely mimic how the brain works, and are supported by arguments from neuroscience, circuit complexity, and machine learning. You can read more about them here:
In New York City, there are a bunch of things happening every weekend.
My girlfriend and I put together a "cheat sheet" with the parties we know about, and our predictions about what will be cool. We then share it with 100 or so of our closest friends friends. They can add more information if they like.
Very convenient! Everyone seems to like this system.
p.s. Social geek with girlfriend? This is not a troll, I promise
The only types of spam I can think of where "following the money" wouldn't work, would be pump-and-dump stock spams -- in that case, the beneficiary could be anyone holding stock in the company.
Actually, there is a very insidious form of spam that we do not (yet) see, where "follow the money" doesn't work: disinformation.
Imagine if shady people on the internet kept whispering rumors and innuendo into your ear. "Britney Spears this... John McCain that..." Sometimes it turns out to be true, so you begin believing that they may have grains of truth. Suddenly there is a whole new component to public discourse on politics, religion, etc.
Let an ultra-intelligent machine be defined as a machine that can far surpass all the intellectual activities of any man however clever. Since the design of machines is one of these intellectual activities, an ultra-intelligent machine could design even better machines
This does not necessarily follow. Here is my analogy:
Once I design a compression algorithm, I will run my compression algorithm on my compressed output ad infinitum!!!
Agreed. The UI of the Motorola Razor is not well thought-out, for example it takes many more keypresses to send a text message than on my old phone (forget the model).
Please don't get motorola. I have had bad experiences with their phones, as have people with whom I have spoken. They emphasize style at the expense of substance. I have found the Razor to be particularly unreliable.
Compression finds the underlying regularities (patterns) in the data.
Here is how it can be used for prediction:
Start with data D. Let c(D) be the size of compressing D. Let's say we want to predict which of two predictions, p or p', is more likely, given D. We say that p is more likely than p' iff c(D + p) c(D + p'), where + is concatenation.
In other words, we predict the statement that is more similar to the previously observed data.
Slashdot Mirror
The PDP books are good, but only after you understand what is currently relevant. I recommend more modern treatments first.
Artificial Intelligence: A Modern Approach by Rusell and Norvig is more or less the standard AI textbook and the book I'd suggest to get an overview of AI and its different methodologies. Mind you, it's over 1000 pages, but a very interesting read.
I reject that. Russell and Norvig really turned me off to AI, because they mainly come from a perspective "good old-fashioned AI" (GOFAI). This mindset is more about discrete symbols and logics, and reject uncertainty, probability, and fuzziness.
I was turned off by the entire field until I began learning about statistical, empirical, and data-driven approaches.
I heartily endorse Bishop (2006). It's a much more modern treatment.
"I have deeply regretted that I did not proceed far enough at least to understand something of the great leading principles of mathematics, for men thus endowed seem to have an extra sense." --- Charles Darwin
Here are some classics in the field. I'll let you google them yourselves.
LeCun et al, 1998: Gradient-based learning applied to handwriting recognition. A deep convolutional net that can read handwriting, and was deployed nationally . Yann LeCun tells me that a patent lawyer in California reimplemented in his free time as a hobby, so it can't be that hard.
LeCun et al, 1998: Efficient BackProp. Tricks and implementation details that are not discussed often.
Btw, as I understand it LeCun was offered a position to be head of Google research. He declined, and Corinna Cortes took the job instead. Regardless, if you googled Yann for a while, Google ads would try to entice you to work at Google.
There is a recent trend in neural networks towards something called Deep Learning. This deep neural networks more closely mimic how the brain works, and are supported by arguments from neuroscience, circuit complexity, and machine learning. You can read more about them here:
Bengio, 2007: Learning deep architectures for AI
In New York City, there are a bunch of things happening every weekend. My girlfriend and I put together a "cheat sheet" with the parties we know about, and our predictions about what will be cool. We then share it with 100 or so of our closest friends friends. They can add more information if they like. Very convenient! Everyone seems to like this system. p.s. Social geek with girlfriend? This is not a troll, I promise
http://www.globus.org/
Obviously, do the exact opposite. Jocks and Cheerleaders.
The only types of spam I can think of where "following the money" wouldn't work, would be pump-and-dump stock spams -- in that case, the beneficiary could be anyone holding stock in the company.
Actually, there is a very insidious form of spam that we do not (yet) see, where "follow the money" doesn't work: disinformation.
Imagine if shady people on the internet kept whispering rumors and innuendo into your ear. "Britney Spears this... John McCain that..." Sometimes it turns out to be true, so you begin believing that they may have grains of truth.
Suddenly there is a whole new component to public discourse on politics, religion, etc.
Let an ultra-intelligent machine be defined as a machine that can far surpass all the intellectual activities of any man however clever. Since the design of machines is one of these intellectual activities, an ultra-intelligent machine could design even better machines This does not necessarily follow. Here is my analogy: Once I design a compression algorithm, I will run my compression algorithm on my compressed output ad infinitum!!!
Agreed. The UI of the Motorola Razor is not well thought-out, for example it takes many more keypresses to send a text message than on my old phone (forget the model).
Please don't get motorola. I have had bad experiences with their phones, as have people with whom I have spoken. They emphasize style at the expense of substance. I have found the Razor to be particularly unreliable.
Compression finds the underlying regularities (patterns) in the data.
Here is how it can be used for prediction:
Start with data D.
Let c(D) be the size of compressing D.
Let's say we want to predict which of two predictions, p or p', is more likely, given D.
We say that p is more likely than p' iff c(D + p) c(D + p'), where + is concatenation.
In other words, we predict the statement that is more similar to the previously observed data.