Editing Machine Learning Meetup Notes: 2009-03-11

Jump to navigation Jump to search
Warning: You are not logged in. Your IP address will be publicly visible if you make any edits. If you log in or create an account, your edits will be attributed to your username, along with other benefits.

The edit can be undone. Please check the comparison below to verify that this is what you want to do, and then publish the changes below to finish undoing the edit.

Latest revision Your text
Line 1: Line 1:
==Machine Learning Meetup Notes: 2009-03-11==
==Machine Learning Meetup Notes: 2009-03-11==
[[Image:IMG_4478.JPG|816px]]


We made perceptrons.  To learn how, you should go to Wikipedia [http://en.wikipedia.org/wiki/Perceptron]
We made perceptrons.  To learn how, you should go to Wikipedia [http://en.wikipedia.org/wiki/Perceptron]
Also reasonably good: [http://www.cs.usyd.edu.au/~irena/ai01/nn/travtute.htm Neural Networks Tutorial], from the Introduction to Part 5.


===Code people wrote===
===Code people wrote===
Line 14: Line 11:
* Ruby by Zhao [https://www.noisebridge.net/wiki/Machine_Learning_Meetup_Notes_Ruby_Zhao]
* Ruby by Zhao [https://www.noisebridge.net/wiki/Machine_Learning_Meetup_Notes_Ruby_Zhao]
* C by Cristian [https://www.noisebridge.net/wiki/User:Cortiz]
* C by Cristian [https://www.noisebridge.net/wiki/User:Cortiz]
* [http://github.com/david415/ml-py/tree/master Python implementation by David Stainton]
* [https://www.noisebridge.net/wiki/davids_perceptron.pl Python implementation by David Stainton]
* [[User:Kaufman/LISP_Perceptron|LISP!]] by John Kaufman
* [[User:Kaufman/LISP_Perceptron|LISP!]] by John Kaufman
* [[User:Ping/Python_Perceptron]] by [[User:Ping]]


Everyone should upload their code!
Everyone should upload their code!
Line 26: Line 22:
Can you build a training set for an OR gate, and train it?  What other operators can you implement this way?  All you need to do is build a new training set and try training, which is pretty awesome if you think about it.  (Hint: What does "separability" mean, and how is it relevant?)
Can you build a training set for an OR gate, and train it?  What other operators can you implement this way?  All you need to do is build a new training set and try training, which is pretty awesome if you think about it.  (Hint: What does "separability" mean, and how is it relevant?)


Let's say we wanted to output smooth values instead of just 0 or 1.  What wouuld you need to change in your evaluation step to get rid of the thresholding?  What wouuld you need to change about learning to allow your neuron to learn smooth functions?  (Hint: in a smooth output function, we want to change the amount of training we do by how far we were off, not just by which direction we were off.)  
Let's say we wanted to output smooth values instead of just 0 or 1.  What wouuld you need to change in your evaluation step to get rid of the thresholding?  What wouuld you need to change about learning to allow your neuron to learn smooth functions?  (Hint: in a smooth output function, we want to change the amount of training we do by how far we were off, not just by which direction we were off.)
 
''One answer: [https://www.noisebridge.net/wiki/Image:NoisebridgeNeuralNetworks_2009MAR17.pdf PDF of Mathematica workspace]''


What if we wanted to do something besides multiple each input by its weight?  What if we wanted to do something crazy, like take the second input, square it, and multiply _that_ by the weight?  That is: what if we wanted to make the output a polynomial equation instead of a linear one, where each input is x^1, x^2, etc, with the weights as their coefficients?  What would need to change in your implementation?  What if we wanted to do even crazier things, like bizarre trig functions?
What if we wanted to do something besides multiple each input by its weight?  What if we wanted to do something crazy, like take the second input, square it, and multiply _that_ by the weight?  That is: what if we wanted to make the output a polynomial equation instead of a linear one, where each input is x^1, x^2, etc, with the weights as their coefficients?  What would need to change in your implementation?  What if we wanted to do even crazier things, like bizarre trig functions?
Please note that all contributions to Noisebridge are considered to be released under the Creative Commons Attribution-NonCommercial-ShareAlike (see Noisebridge:Copyrights for details). If you do not want your writing to be edited mercilessly and redistributed at will, then do not submit it here.
You are also promising us that you wrote this yourself, or copied it from a public domain or similar free resource. Do not submit copyrighted work without permission!

To protect the wiki against automated edit spam, we kindly ask you to solve the following CAPTCHA:

Cancel Editing help (opens in new window)