Tuesday, June 07, 2005

 

Google is your really good friend

Deepak at AI-complete has a nice post about a paper by Vitanyi and Cilibrasi that is causing a buzz. Read the post and read the paper.

Comments:
Oh, Mein Got!! Kolmogorov Complexity!! It remembers me my current professor in Performance Analysis course, he is supervised by the same guy that said once that Soft Computing Approaches are all Gay Things. But anyway, the professor once was talking with me and asking me why I don't pay too much attention with some stuffs in the undergraduation course, meanly with his Performance course. So, at a moment he asked me which are my interests inside the Computation World, I told him that I'm very interested in Evolutionary Computation. He gave me a "precious" advice: Try to mix Genetic Algorithm and Markov Chains. I had, sincerely, a great will to laugh at that moment. But I was polite and just said to him my "Thank You". Because, so far, I understand that Markov Chains consider only the current state of a stochastic process to make a transition to another state. Well, I think that GA's are not so independent and I found once an article about Markov Chains and GA's, the author tried to make an aproximation of a GA using Markov Chains. When dealing with small populations and simple genetic operators, the Markov Chain performed "well", but when the GA's complexity grows up, the Markov Chain performs so poor.

Sometimes that professor seems too much with "The Man Who Only Had a Hammer... and Thought That Everything Around Him Was a Nail".

There is a very interesting quote here in my country: "We do not kick a dead dog and do not throw a stone in a tree that gives no fruit".

See You!!

Nosophorus
 
I am impressed with the hamsters
 
Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?