Return Styles: Pseud0ch, Terminal, Valhalla, NES, Geocities, Blue Moon.

Pages: 1-

GA

Name: Anonymous 2009-05-22 1:07

Dear /prog/,

This summer I need to avoid boredom. I'm teaching myself C, and would like my summer project to be a genetic algorithm. I've done an independent study with a professor, but there was no coding involved, just general theory. What would be a practical problem to find a solution for using a GA for the intention of learning?

Name: Anonymous 2009-05-22 1:27

>>1
need to avoid boredom ... C
Then you've made some bad choices along the way.

Just think of something you'd like to optimize through trial and error.

Name: Anonymous 2009-05-22 1:34

>>2
That's the thing. If it's something simple, I really don't want to go through the hassle. I want something that a human can figure out easily but a computer needs a bit of time to learn.

Hmm, I should look up some of my old Calculus notes for optimization word problems!

Name: Anonymous 2009-05-22 1:35

>>1
READ SICP HURR

Name: Anonymous 2009-05-22 1:36

Some people, when they have a problem with boredom, think, "I know, I'll learn C." Now they have two problems.

Name: Anonymous 2009-05-22 1:43

>>5
Why? It's an extremely portable and widely used program.

Name: Anonymous 2009-05-22 1:44

>>6
And by program, I mean Programming Language

Name: Anonymous 2009-05-22 2:00

done an independent study with a professor
posting on /prog/
implicit knowledge of genetics
[Not knowing] C

IHBT

Name: Anonymous 2009-05-22 2:02

>>4
HURR
Back to the imageboards, please. with this kind of talk.

Name: Anonymous 2009-05-22 2:34

GA MORE LIKE GAY AMIRITE?

Name: Anonymous 2009-05-22 2:36

>>6
So is Java. Somehow both are boring.

Name: Anonymous 2009-05-22 4:47

>>1
Potentially practical and enlightening problems for GAs include:
- Function minimization of a function with multiple minima (f(x,y) = x * sin(4*x) + 1.1 * y * sin(2*y))
- Model fitting (least squares, polynomials etc.)
- The travelling salesman problem
Compare best and mean fitness values for different methods of parent selection (simple, cost-ranking).

Name: Anonymous 2009-05-22 6:05

>>1
I think you should look into finding a linear time factorization algorithm

Name: Anonymous 2009-05-22 7:02

Let it play a game.

Name: Anonymous 2009-05-22 12:06

>>12
Thanks for the ideas.

Name: Anonymous 2011-02-03 3:28

Name: Anonymous 2013-01-19 14:37

/prog/ will be spammed continuously until further notice. we apologize for any inconvenience this may cause.

Don't change these.
Name: Email:
Entire Thread Thread List