Slashdot Mirror

For my master's thesis, I wrote up a GA that evolved Neural Networks that controlled embodied agents in a virtual reality.

Though the scope of GA's and Evolutionary Algorithm's is somewhat limited, within their domain they can be quite powerful, as attested to the video's linked above and the OP. Developing a fluid/lifelike control system for virtual creatures or a system that approximates images without the benefit of Evolutionary Algorithms would likely require much more code.

As a Grad Student, I studied evolutionary algorithms, and my Thesis involved evolving locomotive behaviors in Virtual Agents. While evolved behaviors are interesting, its not surprising that the lying behavior eventually evolved. Evolution will reward behavior that imparts a better chance of survival, and in this case, the lying behavior increased the Agent's chance of survival and replication, therefore it was selected for by the evolutionary algorithms.

The biggest problem with simulated evolutionary systems today is, we don't fully understand the complexity of human behavior, and we are not capable of evolving complex human behavior and mannerisms. It sure would be amazing if we could evolve a Robot or virtual agent to imagine, or do something for fun; however, our models of the Brain are incomplete, and the complexity of the real world too great for us to do such a thing at this moment in time.

Even small scale simulations of the brain, based upon the not so biologically plausible model of a neural network produce Interesting Results. It is likely that a much smaller scale model of the rat brain, when fully understood, would produce intelligent results as well.

Back when I was in University, I did my master's thesis on Embodied Intelligence. I developed a virtual world that adhered to the laws of physics using the ODE physics engine, and within this artificial physical environment I evolved embodied agents. It's quite interesting to watch the videos and see the fluid, almost life-like motions of the evolved behaviors.

I never got around to actually downloading the evolved neural networks into robots, although all my source code is GPL'ed and posted at the above site. So if somebody wanted to evolve their own creatures and download the evolved intelligence into an actual physical robot, it would be interesting to see the results

It is amazing to observe the progress of evolution. Initially, the agents act as though they don't know what they are doing, their movements being very laboured and imprecise. As the population continues to evolve, individuals from the later generations begin to increasingly show signs of intelligent improvements. Such as being able to coordinate and time their movements to afford them better walking/running behaviors. What is really surprising is that as I allowed the population to continue to evolve, I saw behaviors that I could never have predicted. Some of the agents began to do "tricks" and evolved some strange jumping or sommersalting behaviors, not unlike that of a gymnast.

If you get a chance, check out my thesis. It is freely available (with GPL'd source code) at: http://www.erachampion.com/ai

It is amazing to observe the progress of evolution. Initially, the agents act as though they don't know what they are doing, their movements being very laboured and imprecise. As the population continues to evolve, individuals from the later generations begin to increasingly show signs of intelligent improvements. Such as being able to coordinate and time their movements to afford them better walking/running behaviors. What is really surprising is that as I allowed the population to continue to evolve, I saw behaviors that I could never have predicted. Some of the agents began to do "tricks" and evolved some strange jumping or sommersalting behaviors, not unlike that of a gymnast.

If you get a chance, check out my thesis. It is freely available (with GPL'd source code) at: http://www.erachampion.com/ai

Check out this site: AI

I have run apache/php on my Zaurus and it works great for serving web pages across 802.11b! I don't know if anyone would try this setup as a production web server; however, it's cool that it works.

Guru
ERA Champion Real Estate

Eventually, the laws of physics will catch up to us, and Moore's law will not be upheld. How long this will take is anyone's guess; however, many respected scientists believe that uniprocessor systems will continue to see advances into 2006.

At some point in the future, we will not be able to make the die sizes any smaller (right now we are at 0.13 micron) due to the quantum effects that such small circuits display, engineers will be forced to seek alternative methods of performance enhancement. In addition, I suspect, there will also come a time when we can not get a transistor to switch any faster (you can't switch in 0 time). Then what?

So far, the only alternative is Parallel Distributed computing. Having a cheap/efficient parallel computer based on a commodity processor is where the future will be. Operating System support and Compiler support is another story, but Mosix for linux is a good start! =)

Guru
ERA Champion R/E, Inc.

Most new RPTV's and HDTV sets can support a 480p input. If the XBox truely outputs a 480p signal, I would expect the quality to be superior to that of a Playstation 2 who's output is 480i which can then be converted to 480p by the TV set.

Though you end up with 480p in both cases, its quite obvious the XBox solution is the optimal one.

Gene Ruebsamen
Orange County Real Estate

1) TogetherSoft allows for UML->Source and Source->UML. Try the Community version of Togethersoft for FREE, only the print functionality has been disabled in this version. Has great syntax highlighting, and is fairly fast for a full blown java application.
http://www.togethercommunity.com/community-edition .shtml

2) Forte for Java CE the Community Edition of the popular Forte Software is what I personally use for my Java Development. While not having the UML features of the Togethersoft product, I find it suits my needs just fine. An integrated debugger, and syntax highlighting make for a friendly (and free) product!
http://java.sun.com

3) Glimmer for linux. While not a full blown IDE, for simple projects, I find myself using glimmer alot. It's quick, written in C, and supports syntax highlighting for a plethora of languages (php, c, c++, java, perl, and lots more!), and best of all.. it's GPL!
http://glimmer.sourceforge.net

Gene Ruebsamen
Orange County Real Estate

A Simple Genetic Algorithm:
1) Randomly create population of individuals and their genome.
2) Calculate fitness of each individual
3) Probabilistically select individuals for mating, based upon their fitness scores.
4) During Mating, use crossover (usually 70% chance) and mutation (usually .1 - .01% chance) on the chromosome(genome) to create a new offspring population.
5) If done, then finish, otherwise goto step 2 with the new population.

Using the Genetic Algorithm to evolve new crossover operators seems feasible; however, evolving the mutation operator to produce individuals of higher fitness is flawed logic. Mutation's purpose is to prevent genetic code from becoming "lost" (ie sometimes when you lose genetic information from crossover, you can never get it back w/o random mutations). By evolving the mutation operator to produce individuals of higher fitness, I suspect you will run into the Local Optima problem, where you've reached a local optimum fitness, but you are nowhere the global optimum fitness for the population.

Mutation is very important in preventing the local optimum problem, changing the way mutation works will probably have a detrimental effect on the performance.

Gururuse
ERA Champion Realty, Inc.

However, I don't see how any American law would affect Bin Laden in Afganistan? Committing terrorist acts is ALREADY illegal here in the US. What effect is passing another hundred laws going to have, other than reducing the freedoms of law abiding citizens?

If congress banned encryption, it would only affect law abiding Americans, and would have no effect on outlaws or criminals.

Guru
ERA Champion Real Estate

The fallacy with banning encryption is if you outlaw encryption, only outlaws will use encryption!

Gene Ruebsamen
Orange County Real Estate
ERA Champion Realty, Inc.

My employer is ERA Champion Realty, Inc., a Real Estate company based in Orange County, California with about three dozen employees.

Our manager has committed to donating $500 for each home sale made during the next 60 days. Counting the sales we already have scheduled to close this month, we are close to the $5000 mark.

Just wanted to let you all know that some of the smaller companies are pitching in too!

* Seems faster than the old 0.6 or 0.7 builds. (ie the menus seem zippier).
* Seems to load faster than previous versions.
* Still can't minimize the download window w/o minimizing the entire Mozilla app.... (what gives?)
* Still having trouble installing JRE 1.3
Overall, 0.8 seems to be faster, and generally better than any of the previous builds.
Gururise
Garden Grove Real Estate

I'm using it now to type this message. My general impressions are it's a bit more polished that 7.1 was; however, my first install failed, after going to the 2nd cd, it just froze on Penguin Command (while installing). After a reboot, and re-install (unselecting Penguin Command to be safe) everything worked great. Had to play with the network settings a bit after installation to get the default gateway set properly, other than that, everything seems to work fine.

Gene

www.erachampion.com