[Cs254f11] My plan

[Lee Spector]

Consider the symbolic regression problem in Chapter 4 of the book (trying to evolve x^2 + x + 1).

Now think about doing the same thing but after:

- adding some additional functions that return numbers derived from MIDI files; these can take numeric parameters (which might specify locations in the file or kinds of data or whatever)

- evaluating error not by comparing to x^2 + x +1 but rather by comparing to (def josiahs-judgements {:file1 0.8, :file2 0.3, ...})

Make sense?

 -Lee

On Oct 18, 2011, at 11:34 AM, Wm. Josiah Erikson wrote:

> Or it could be just me who doesn't have a solid handle on it yet :)
> 
> I sortof feel like I get the generic case, but I'm starting to get more of a feel for applying it to specific cases.
> 
>    -Josiah
> 
> 
> On 10/18/11 11:27 AM, Lee Spector wrote:
>> On Oct 18, 2011, at 11:00 AM, Thomas Helmuth wrote:
>>> P.S. Lee: This thread is making me wonder if there is some general confusion in the class about actually evolving programs. Maybe we should discuss this more in class?
>> Yep, good suggestion, and in line with what I'm planning -- moving from the evolvesum thing that I showed yesterday to GP proper, in the context of the new reading assignment from the Field Guide. I'm also hoping that everyone's focus on their projects, which should now be firmed up pretty well, will help them to focus on the specifics that they'll need to proceed.
>> 
>>  -Lee
>> 
>> --
>> Lee Spector, Professor of Computer Science
>> Cognitive Science, Hampshire College
>> 893 West Street, Amherst, MA 01002-3359
>> lspector at hampshire.edu, http://hampshire.edu/lspector/
>> Phone: 413-559-5352, Fax: 413-559-5438
>> 
> 
> -- 
> Wm. Josiah Erikson
> Network Engineer
> Hampshire College
> Amherst, MA 01002
> (413) 559-6091
> 

--
Lee Spector, Professor of Computer Science
Cognitive Science, Hampshire College
893 West Street, Amherst, MA 01002-3359
lspector at hampshire.edu, http://hampshire.edu/lspector/
Phone: 413-559-5352, Fax: 413-559-5438