addressalign-toparrow-leftarrow-rightbackbellblockcalendarcameraccwcheckchevron-downchevron-leftchevron-rightchevron-small-downchevron-small-leftchevron-small-rightchevron-small-upchevron-upcircle-with-checkcircle-with-crosscircle-with-pluscontroller-playcredit-cardcrossdots-three-verticaleditemptyheartexporteye-with-lineeyefacebookfolderfullheartglobe--smallglobegmailgooglegroupshelp-with-circleimageimagesinstagramFill 1launch-new-window--smalllight-bulblinklocation-pinm-swarmSearchmailmessagesminusmoremuplabelShape 3 + Rectangle 1ShapeoutlookpersonJoin Group on CardStartprice-ribbonprintShapeShapeShapeShapeImported LayersImported LayersImported Layersshieldstartickettrashtriangle-downtriangle-uptwitteruserwarningyahoo

RE: [Cleveland-AI-ML-support-group] Languages/Machine Learning Math

From: user 1.
Sent on: Thursday, August 25, 2011 11:07 AM

Yikes! That’s a lot to digest!

 

From: [address removed] [mailto:[address removed]] On Behalf Of Joe
Sent: Thursday, August 25,[masked]:03 AM
To: [address removed]
Subject: Re: [Cleveland-AI-ML-support-group] Languages/Machine Learning Math

 

 Your welcome and thanks for hosting the meeting. I'll check that book out, the more 'automatic' we get with the math, the easier we will be able to understand and employ the AI ML algorithms. The ML approach seems to be automated function discovery and fitting with algorithms using those math methods. That is exciting, because that also seems to be the core part of the successful Eureqa AI system:
http://creativemachines.cornell.edu/eureqa

 I think 2 hours will be plenty for the first meeting. I for one haven't taken a college course in years, so it will be useful for me to build up to the needed college level of study and exercise solving before the course starts. For courses of this type, that's probably 1 to 2 hours of study and exercise solving a day for each separately, in addition to watching the lectures.

 For this week on the ML course we could cover sections 1 and 2 of the Linear Algebra review. That is roughly equal to the first 21 Khan Linear Algebra videos. After the first 21 videos a lot of the remaining videos for the rest of the course can probably be skipped, I'll keep a list on the wiki. The only needed thing remaining for the first ML study session, is some problems to have worked on for discussion. Maybe we can each make up our own problems as variations on what Khan discusses.

Joe



--- On Thu, 8/25/11, Timmy Wilson <[address removed]> wrote:


From: Timmy Wilson <[address removed]>
Subject: Re: [Cleveland-AI-ML-support-group] Languages/Machine Learning Math
To: [address removed]
Date: Thursday, August 25, 2011, 10:09 AM


Awesome!!

Thanks for the info Joe.

I have to leave @ 12:00 on Sat. 

Is 2 hours enough time for our first meeting?

If you want something related to probability -- Stephen Gould's 'Full House' is a nice read -- especially if you're in to baseball

You'll learn about distributions, standard deviations, tails, walls, ect.. -- he puts it all in a 'naturalist' context

http://www.amazon.com/Full-House-Spread-Excellence-Darwin/dp/0609801406



On Thu, Aug 25, 2011 at 1:47 AM, Joe <[address removed]> wrote:

 Norvig had originally written running sample code for the book in Lisp, than rewrote it to Python after getting a job at the Python using Google, imo, although he characterizes that change differently:
http://norvig.com/python-lisp.html

 The book uses a Python like none running pseudo code for most of it's quizzes and their answers, so translating it as necessary and running it in other languages would be extra work. However if any one wants to do that for language X in addition, after covering the quizzes in the books pseudo code, that would be pretty cool to take a look at. I'll be trying that in Lisp on my own for those that are interested.

 After reading the first lecture and problem set for the Machine Learning course I realized something, that without an extensive review of Linear Algebra and Probability theory, it's going to be significantly difficult bordering on impossible for none recent math graduates to keep up with this course. So I suggest that we devote the time before the machine learning course to that review. For this first week we could cover the 'Linear Algebra Review and Reference' pdf linked here:
http://www.stanford.edu/class/cs229/materials.html

These videos are also a great supporting resource:
http://www.khanacademy.org/?video=introduction-to-matrices#linear-algebra

The ML course essentially appears to be a 'Linear Algebra/Probability Theory algorithms for AI' course. So getting pretty sharp in those areas would seem to be required for this course.

Regards,
Joe




--- On Wed, 8/24/11, Aki Iskandar <[address removed]> wrote:


From: Aki Iskandar <[address removed]>
Subject: Re: [Cleveland-AI-ML-support-group] When to meet
To: [address removed]
Date: Wednesday, August 24, 2011, 9:07 PM

Thanks Joe - for your contribution to this alias, and setting up a Wiki.

Gang - here is some fodder on languages

Languages:  Yes - LISP is excellent - but I believe that Peter Norvig teaches his students using Python as the language ... because it is much easier than LISP (not everyone can pick LISP up without a tremendous amount of effort), and fairly expressive.

Having said that, languages don't matter too much to me.  I prefer Python - but can use Clojure (a LISP variant) or Objective-C ... which is actually a very nice language with expressive power as well.   Lately, I'm focusing on Python and Objective-C ... but can use Ruby, or straight LISP (with a bit of a stretch ... it's not that straight forward for most - including me) ... whatever people want to use is ok by me.

Thanks,
~Aki


On Wed, Aug 24, 2011 at 1:22 PM, Joe <[address removed]> wrote:

 Sounds Great. See you all on Saturday at 10. I created a wiki for the study group:

http://freedombluesky.com/aiml

User: aiml
password: square

To edit or add pages create an account.

Micheal asked what languages would be used. The book uses pseudo code. I will be translating that into Lisp for additional study. I'm a Lisp evangelist because I believe it's the most concisely self programmable programming language. Lisp is used in the book 'Structure and Interpretation of computer programming' which has been called one of the most important programming books by Peter Norvig. Unfortunately that book can be long and challenging to work through, but one can learn similar material, and Lisp, with this much faster and easier book, 'Concrete Abstractions': https://gustavus.edu/+max/concrete-abstractions.html

Paul Graham is one of the most famous and successful startup programmers and incubator directors, and is a strong proponent of Lisp, the original AI language:
http://paulgraham.com/icad.html

Regards,
Joe



--- On Wed, 8/24/11, Timmy Wilson <[address removed]> wrote:


From: Timmy Wilson <[address removed]>


Subject: Re: [Cleveland-AI-ML-support-group] When to meet
To: [address removed]

Date: Wednesday, August 24, 2011, 9:45 AM

 CrazeThe votes were pretty evenly distributed


3 morn
3 afternoon
2 night

Let's go w/ sat @ 10am




--
Please Note: If you hit "REPLY", your message will be sent to everyone on this mailing list ([address removed])
This message was sent by Joe ([address removed]) from Cleveland AI + ML support group.
To learn more about Joe, visit his/her member profile
To unsubscribe or to update your mailing list settings, click here

Meetup, PO Box 4668 #37895 New York, New York[masked] | [address removed]




--
Aki Iskandar
[address removed]
(440)[masked]

"The past does not equal the future.  You have a blank slate upon which to plan, and build, the life you want."





--
Please Note: If you hit "REPLY", your message will be sent to everyone on this mailing list ([address removed])
This message was sent by Aki Iskandar ([address removed]) from Cleveland AI + ML support group.
To learn more about Aki Iskandar, visit his/her member profile
To unsubscribe or to update your mailing list settings, click here

Meetup, PO Box 4668 #37895 New York, New York[masked] | [address removed]





--
Please Note: If you hit "REPLY", your message will be sent to everyone on this mailing list ([address removed])
This message was sent by Joe ([address removed]) from Cleveland AI + ML support group.
To learn more about Joe, visit his/her member profile
To unsubscribe or to update your mailing list settings, click here

Meetup, PO Box 4668 #37895 New York, New York[masked] | [address removed]






--
Please Note: If you hit "REPLY", your message will be sent to everyone on this mailing list ([address removed])
This message was sent by Timmy Wilson ([address removed]) from Cleveland AI + ML support group.
To learn more about Timmy Wilson, visit his/her member profile
To unsubscribe or to update your mailing list settings, click here

Meetup, PO Box 4668 #37895 New York, New York[masked] | [address removed]





--
Please Note: If you hit "REPLY", your message will be sent to everyone on this mailing list ([address removed])
This message was sent by Joe ([address removed]) from Cleveland AI + ML support group.
To learn more about Joe, visit his/her member profile
To unsubscribe or to update your mailing list settings, click here

Meetup, PO Box 4668 #37895 New York, New York[masked] | [address removed]

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy