Togelius At New Scientist And Slashdot

Our blog friend Julian Togelius has a nice post upon a New Scientist's article which relates his research on car control. Slashdot also has a thread about the same subject.

Happy Birthday To Genetic Argonaut!!!!!

Last Tuesday was accomplished the second consecutive year that I keep on-line this humble and, as someone said elsewhere, relatively obscure blog. :)

By the way, I am happy with what has happened from last two years until now and I have found out so many interesting, nice, and friendly persons inside the so-called blogosphere.

It has been a long journey since my first evolutionary computation steps and I realize how many important things I have learned.

Surely I also found interesting and nice persons from academia - even though only via e-mail (at least all those outside Brazil)! :)

Here you are a little thank list:

Eurípedes Pinheiro dos Santos - The person who introduced me to this crazy and chaotic EC world! Thank you for all the coffee, A.I. papers, books, comments, conversations, critiques upon EC ("Genetic Algorithms are not a panacea!"), advices, help, explanations, funny moments, and your worthful friendship. :)

Hans-Paul Schwefel - Thank you for all advices (see below), helpful and (always) intelligent comments, your interesting critiques on the first version of the Evolutionary Computation Classics - Volume I, sensible suggestion about optimization in general, all the A.I./EC papers you sent me from Germany, and some funny e-mail moments - remember the Portuguese Space Program and the bumblebee jokes! :)

"My advice to young men (and women, of course) is always, NOT
to follow the advice of old men (and women). Just follow
your intuition after looking carefully around, taking into
consideration your skills and, above all, your curiosity and fun.
I never followed the advice of elder people, just tried to
increase the number of options I could get. What I earned this
way, was not what I aimed at (I wanted to become an astronaut just
after Sputnik I went into the orbit), but nevertheless I liked
how things were developing (of course, I tried to make things
happen that way, a bit)."

HPS


Antje Schwefel - Thank you for delivering my (skunk) texts of the Evolutionary Computation Classics - Volume I to your husband, the Power Point presentations you sent me via e-mail, the translation of my name to Chinese (see below), the funny moments we had while we were trying to translate my name to Chinese (I feel it difficult as a German to explain Chinese to a Brazilian using English!!!), and for your patience with me after I sent 50 (!!) questions to your husband to answer! :)












Marcelo = "The guy, who can hug a horse!"

Augusto = "This beginning is meaningful, based on history!"

Rasmus Ursem - For your helpful comments and explanations while I was implementing a Diversity Guided Genetic Algorithm (DGGA) - see here.



The DGGA (or DGEA - Diversity Guided Evolutionary Algorithms) is an interesting idea. Instead of using the operators of selection, mutation and crossover through the usual manner (such as in the Simple Genetic Algorithm), the DGEA uses a diversity measurement (see above) to switch between phases of exploitation - when selection and crossover (diversity decreasing operators) are more active and should exploit promising solutions - and exploration - when mutation (diversity increasing operator) is more active and should increase the diversity. Theoretically, the DGEA should be able to escape local optima because the operators will force higher diversity regardless of fitness.

It's a pity that I did not post anything about the DGEA I wrote and the comparison I made among a DGEA, a GA, and an ES. The ES was more robust than the other two.

Xin Yao - Thank you for replying me (via e-mail) some questions about evolutionary computation in general.

Zbigniew Michalewicz - Thank you for replying me (also through e-mail) some questions about evolutionary computation in general.

David E. Goldberg - Thank you for the argumentation upon genetic algorithms and patents. Thank you also for linking my blog from yours.

Amir massoud Farahmand - Thank you for your friendly comments at my blog and, also, for the nice blog interaction we have had. :)

Julian Togelius - Thank you for all those nice and interesting videos upon car control through artificial intelligence! It was nice to see that I am not the only guy in the blogosphere who uses Evolution Strategies (ES) - Julian uses ES to evolve neural networks. Keep posting cool stuffs like those! :)

Damien François - Thank you for linking my blog from yours and, also, for the post you made upon my posting on evolving lens through evolutionary algorithms.

The Link Index - Thank you for linking my blog from your blogs!

Thank you to all those friends who visit, link to, and read this simple blog! :)

Até Mais!!! :)

Machine Learning Videos

Machine Learning Thoughts have had an important and nice initiative: Set up video resources upon machine learning available on the web. See here and here.

There are interesting videos, such as Winning the DARPA Grand Challenge.

Feel free to suggest some video. :)

Evolving Telecommunications Through Simulated Evolution

MEDAL Blogging has an interesting post upon antenna design through simulated evolution.

That post reminded me of three other similar works:

William Comisky, Jessen Yu, and John R. Koza Genetic Programming approach to evolve antenna: Automatic Synthesis of a Wire Antenna Using Genetic Programming.

Scott Santarelli, Tian-Li Yu, David E. Goldberg, Edward Altshuler, Teresa O'Donnell, Hugh Southall, and Robert Mailloux work upon military antenna design via Simple Genetic Algorithm (SGA) and Estimation Of Distribution Algorithms (EDA). Their work shows us the performance of a SGA and an EDA on antenna (automatic) design. The EDA exhibits a considerable better performance than the SGA - Military Antenna Design Using Simple and Competent Genetic Algorithms.

NASA Evolvable Hardware System Group work on satellite antenna design.

The first time I saw those NASA antennas, I got atonished how different they are from more typical antennas, such as the Yugi-Uda or the parabolic dish. See here an old post of mine about it.

Marvin Minsky At SETI Radio Network

The famous computer scientist Marvin Minsky will be - on the next March 15, 2007 - at SETI Radio Network giving a talk upon consciousness together with

Dr. Seth Shostak - SETI Institute - Senior Astronomer;
Patricia S. Churchland - Professor of Philosphy, UCSD;
Jaron Lanier - Professor of Philosophy, University of California, Berkeley;
John R. Searle - Professor of Philosophy, University of California, Berkeley;
Karla Heidelberg – Assistant Professor, Department of Biology, University of Southern California.

So, tune your radio, or browser, to linsten Minsky and the other guests talking about consciousness.

How to listen? Here you are some instructions:

How To Listen.

A FAQ: Frequently Asked Questions.

Marvin Minsky's View Upon Genetic Algorithms

I found an old and interesting post from the Grey Thumb blog in which the famous computer scientist and artificial intelligence pioneer, Marvin Minsky, gives his opinions on the current state of artificial intelligence, the progress it have made, and some critiques about some "non-traditional" A.I. approaches, such as genetic algorithms, genetic programming, and neural nets.

Below there is an excerpt from his lecture: It's 2001, where's HAL? - In audio format: here (to play it on your audioplayer) or here (to download the mp3 file). Here you are the excerpt:

"Genetic algorithms are very popular. I can't figure out why because, in almost all respects, they are worse than the traditional artificial intelligence heuristic search. What genetic algorithms do is use the computers ten thousand times faster to make lots of things to try. Then you have a competition so that the ones that succeed better in solving some problem, or faster, replace the ones that took longer. However, in real life I think that's the wrong thing. And evolution itself is screwy about this. The important thing is not to remember what led to success - or half of the thing you should remember is what worked. The other half is what are the 100 most common mistakes. When I was training for a Ph.D. in mathematics, everyone understood that in the mathematics world. If you hear a theorem, then you also want to know the 10 most likely ways that it won't apply. Of course a theorem is always true if the conditions are true, but if this was true for a compact set, is it also true for a locally compact set, and if not where is the counter example that shows why that kind of reasoning breaks down.

What evolution and genetic algorithms don't do -tell me if I'm wrong- is keep any record of why all those poor losers died. If it weren't for a almost religious, superstitious worship of imitating genetics which took 600 million years, well, [] 400 million years, to get to us from fish... You could say, boy, if [we] had kept some records of what went wrong and spent about the same amount of energy on learning how to avoid bugs, maybe it would have taken only 5 million years instead of 400? Who knows? But I think people who look at genetic algorithms and don't notice that they don't solve any problems that require deep thought should... I could go on all day?"

Oh! What a good hearted words!

It's an extremist statement. The most extremist part of his speech on genetic algorithms (GA) is the following:

"Genetic algorithms are very popular, and I can't figure out why, because in almost all respects they are worse than the traditional artificial intelligence heuristic search."

Surely, it is a strongly biased opinion, since Minsky represents the A.I. old school branch.

I wonder if subjects such as evolvable hardware would be possible through traditional heuristic search. By the way, taking into account the speech's title - It's 2001, where's HAL? - would not those traditional heuristic search methods have something to do with? Let alone the fact that, along the 1960s, there was an over-optimistic view on computers and what they would soon do in the next 20 or 30 years. Then, what did happen after 20/30 years? In a single word: Deception!

It's interesting (and strange) how Minsky has so many misconceptions upon genetic algorithms (and evolutionary computation too).

Thanks Heavens that Minsky (maybe) does not know what is a decomposable/separable function!

-------------------------------------

Also via Intelligent Machines.

-------------------------------------

Intelligent Machines Blogs About Genetic Argonaut

Intelligent Machines has a post on my posting upon Evolving Lens Through Evolution Strategy And Genetic Algorithm.

Intelligent Machines is an interesting blog upon Artificial Intelligence (A.I.). Take a moment and read it! :)

Thank you, friend, for citing this blog! :)

P.S: I have an old post related to evolving lens through evolutionary algorithms:

Evolution Of An Optical Lens Through Evolution Strategy

Michael Trick’s Operations Research Blog Blogs About Genetic Argonaut

Michael Trick’s Operations Research Blog has a post upon my last posting on Evolving Lens Through Evolution Strategy And Genetic Algorithm. I am very grateful for being cited. :)

And to greet back his friendly post, Michael's blog is the first one to be included on a new category of links in this blog: Operations Research. :)

Thank you, friend, for your citation and I hope we can exchange interesting and friendly ideas upon optimization. :)