Forgot your password?
typodupeerror
Programming

Journal: Mario AI Competition

Journal by jtogel
We're running a competition to see who can program the best AI for a version of Super Mario Bros. It's about each time step deciding what to do - run, jump, shoot etc. - based on a description of the platforms, items and enemies around Mario.

This is hard. So hard we believe that some sort of machine learning algorithm will be necessary to reach good playing performance. But really, any approach is game. We welcome hard-coded submissions, and we welcome commercial AI programmers, academics and amateurs alike. Whoever wins (maybe you?) this will be really interesting.

The competition is associated with two IEEE conferences and there are cash prizes available for the best submissions.
Classic Games (Games)

Journal: Can we design fun games automatically?

Journal by jtogel
What makes games fun? Some (e.g. Raph Koster) claim that is fun is learning - fun games are those which are easy to learn, but hard to master, with a long and smooth learning curve. I think we can create fun game rules automatically through measuring their learnability. In a recent experiment we do this using evolutionary computation, and create some simple Pacman-like new games completely without human intervention! Maybe this is the future for game design? (Blog post, paper).
Robotics

Journal: Mini-Grand Challenge Organized by UK University

Journal by jtogel
At Essex, we have for some time been working on automatically learning how to race cars in simulation. It turns out that a combination of evolutionary algorithms and neural networks can learn how to beat all humans in racing games, and also come up with some quite interesting, novel behaviours, which might one day make their way into commercial racing games. While this is simulation, the race is now on for the real thing - we are setting up a competition for AI developers, where the goal is to win a race between model cars on real tracks. As the cars will be around half a meter long, the cost of participating will be a fraction of that for the famous DARPA Grand Challenge, whereas the challenges will be similar in terms of computer vision and AI.
Robotics

Journal: AI swarm helicopters teach themselves how to fly

Journal by jtogel
Media has previously reported on the Ultraswarms project, aiming to create swarms of Linux-powered miniature helicopters with a hive mind. For those who wonder how the project is coming along, I can report that we have been able to use artificial evolution to automatically create neural networks to control the helicopters. In other words, the helicopters learn to fly all by themselves through trial-and-error, which is useful as they are pretty darn hard to control manually. In fact, the evolved networks work better than the best human-made solutions. Work continues...
Slashdot.org

Journal: Why do I keep reading Digg when Slashdot is better?

Journal by jtogel

They say Digg is bigger than Slashdot than these days. Bigger, better, newer, 2.0 and what not.

Indeed, I find myself scanning the Digg front page more often than Slashdot's ditto. Far too often, actually. Obsessively often? Well well, I actually get proper work done now and then.

Still, it's simply not true that the quality of the "news" on Digg's frontpage is higher than the news on Slashdot. On the contrary, Digg suffers from a disgusting amount of mob mentality. Half of what's on the front page is not news at all, merely short text snippets propagating rumours that everybody's already heard or views that most people already agree with. Which is why they get digged to the front page: people like to hear/read what they already believe in.

And this is not really a problem with Digg, it's a problem with people. (And like all problems we can't really do anything about, it's not even worth thinking of it as a problem, just as a fact.)

Slashdot, on the other hand, has sometimes-competent editors that sometimes put an effort into selecting and editing stories. While I might learn something I didn't already know from either Digg or Slashdot, I'm far more likely to learn something new about something I didn't already know anything about from Slashdot. Crucial difference.

So, back to the question: Why do I keep reading Digg when Slashdot is better? Because the Digg is updated more often, and the items are shorter. It's that simple. I think.

User Journal

Journal: AI: All fun and games

Journal by jtogel
Who believes in artificial intelligence nowadays? Not many, it seems. Some say that if we were so simple that we could understand ourselves, we would be so stupid that we couldn't. So we have to make the AI construct itself! That's the topic of my latest blog post, where I also claim that the secret lies in using computer games to do this.
PC Games (Games)

Journal: What makes racing fun?

Journal by jtogel
Me and Renzo are working on a paper on how to automatically create fun racing tracks. And not only that, but racing tracks that are fun just for you - which means that we must first model how you drive with a neural network, and then use this model of your driving to create the tracks.

Grand plans, but will it work? Well, we'll see soon. In the meantime, here's a question for all of you: what exactly is it that makes a racing game fun? And what is it that makes a particular track/circuit in racing game fun?

Of course, in the end we want some quantitative measure we can put into our code, but the quantitative part is really up to us. What we're looking for right now is suggestions, ideas. Any sort. What makes racing fun? Please leave your suggestion below, or in the comments to this blog post.
Robotics

Journal: Recent advances in car racing AI

Journal by jtogel
With all the attention (rightly) given to the DARPA Grand challenge(s), it is a bit surprising that current car racing AI is so sloppy. Computer-controlled cars typically drive according to predefined racing lines and don't learn or adapt their driving styles to your driving. One notable exception is Microsoft's inventive AI in Forza Motorsport, where you can train "drivatars" to drive just like you.

At the University of Essex we are instead working on using evolutionary computation (such as genetic algorithms) and neural networks to control cars. The objective is to have the networks learning robust competitive driving behaviour without human input, and further to continue refining its behaviour when facing new tracks and opponents. So far we have come pretty far in the case of a single car on a track, and we are making progress on the two-car case. More details here, including videos and original papers.

A penny saved is a penny to squander. -- Ambrose Bierce

Working...