Follow Slashdot stories on Twitter


Forgot your password?
DEAL: For $25 - Add A Second Phone Number To Your Smartphone for life! Use promo code SLASHDOT25. Also, Slashdot's Facebook page has a chat bot now. Message it for stories and more. Check out the new SourceForge HTML5 Internet speed test! ×

Submission + - How Darwin plays StarCraft (

jtogel writes: Today it became official that StarCraft will become the next big AI benchmark, as Blizzard and Google DeepMind announced a joint effort to build an API for the game.
We show that we can use evolutionary computation to play StarCraft micro by selecting among scripts at runtime. Not only does this work, but it plays better than any competing algorithm. This is one of the first practical uses of evolutionary computation to select actions in real time.

Submission + - Student Starcraft AI Tournament 2015 Finals 20:00 CET tonight (

An anonymous reader writes: The finals of the biggest tournament for starcraft broodwar AIs will be held tonight. First the semifinals, with Killerbot vs Stone and ZZZKbot vs XIMP, then the 3/4 place match, a man vs machine match, then the grand finals.

This is not about hacking, but about mostly students doing AI research using broodwar as a platform. The games are awsome and as the most complex AI challenge of the now, this scene is highly relevant to all tech lovers.

Submission + - What if videogames had actual AI?

jtogel writes: It is commonly agreed that game AI in (commercial) games is a completely different beast than AI that is being research in academia. This article explains the difference very well and discusses some of the reasons for it. However, what would we be able to do in videogames if we had access to actual working AI? This post describes a vision for games that has actual AI in them.

Submission + - Lara Croft explores her players through data mining

jtogel writes: Whenever you play a game of Tomb Raider: Underworld, a comprehensive record of your playing activities is collected on servers at Square Enix. Pretty much everything is tracked: from number of deaths, causes of death, requests for help, total and relative play time and rewards collected. Researchers at the University of Bonn, Fraunhofer IAIS and Northeastern University have mined this data to identify how playing behavior evolves throughout the entire game.
Using unsupervised behavioral clustering algorithms on gameplay data from 62,000 players, they identified six archetypes that both offered explanatory strength and representation value difference. Confirming earlier work that clustered players into Runners, Pacifists, Solvers and Veterans, this research found consistent spread of behavior at all levels of the game except when the design of a level enforced defined play attitudes. What’s more, playing styles vary and evolve as you play the game. This research helps game designers identify how players change from one type of behavior to the other, for example move from novice to expert, or from a non-paying user to become a paying user. (So that they can put all their effort into the ones that will eventually pay?)

Comment Re:Yavalath (Score 1) 112

Yes, Cameron's work was one of our sources of inspiration. The Ludi system that produced Yavalath is clearly a milestone in research on automatic game design. Another source of inspiration was my own work on automatic game design for simple PacMan-like games, which was carried out at the same time as Cameron's work and is described in this paper:
In general, this line of research is still in its infancy, as we are trying to figure out new ways of evaluating game quality and representing various aspects of games.

Comment Some additional info (Score 1) 112

The original papers describing the work can be found here:

Similar evolutionary techniques have been used to generate a number of different types of game content, including Starcraft maps, Super Mario levels, rocks, dungeons, weapons... Here's an overview:

Submission + - AI system invents new card games (for humans) (

jtogel writes: This New Scientist article describes our AI system that automatically generates card games. The article contains a description of a playable card game generated by our system. But card games are just the beginning... The card game generator is a part of a larger project to automatise all of game development using artificial intelligence methods — we're also working on level generation for a variety of different games, and on rule generation for simple arcade-like games.

Submission + - How does modularity evolve? ( 1

JimmyQS writes: "As programmers know, modularity is critical to making reusable, adaptable software. However, modularity is not instinctive for beginners and must be learned via painful training. Biology faces a similar problem: modularity is useful to make species more adaptable, but how did it evolve in the first place? Surprisingly, computational simulations of 25,000 generations of evolution reveal that modularity does not evolve because it makes organisms more adaptable. Instead, modularity evolves as a by-product for selection to reduce the "wiring costs" of a network. The discovery greatly advances research into evolving artificially intelligent robots, a field where the inability to evolve modular designs has long been thought to be a key roadblock to evolving truly complex, intelligent neural networks.

The paper was published in the Proceedings of the Royal Society. You can also watch modularity evolve in this video."


Submission + - Create your own Monopoly to understand the world around you (

jtogel writes: "Can we use games to help people explore the massive amounts of data that are increasingly available to everyone? In a recent paper, which New Scientist write about here we describe a program that can generate Monopoly Boards from open data streams. The player selects what is important to him or her (child poverty? education? median income?) and the program uses AI methods to create a new game board. The full paper is available here."

Submission + - Human intelligence is overrated (

jtogel writes: "People often wonder when computers will become as intelligent as humans. This question assumes that being as intelligent as a human is a worthy goal for an artificial system. But really, are humans all that smart? If they are not, this has implications for how we should do AI research."

Submission + - Girl uploads photo of money to Facebook, Family robbed (

mask.of.sanity writes: Robbers armed with a knife and a wooden bat attacked an Australian family after their 17 year-old daughter uploaded a photo of cash to Facebook.

The daughter lived in Sydney some 150 kilometres away where she took the photo while helping to count her grandmother's life savings.

It was unknown how much money was counted or how the thieves obtained her family's home address.

Comment Important clarification (Score 1) 114

The aim of the competition is not merely to create random levels, but to generate levels that are tailored to particular players' playing styles and skills. The level generator gets as inputs how well a player performed on a test level and various metrics detailing e.g. how much time the player spent jumping and running and how many fireballs were fired. The level will then be judged by the player who played the test level. See more about the rules at: Also, please note that those who submit a level generator can also submit a paper about it to the CIG conference.

Submission + - New AI challenge is all about wanton destruction (

togelius writes: "Previous years have seen a number of car racing competitions where neural nets, rule-based systems and other fancy AI techniques have been put to the test by letting them drive on a track and seeing who gets the best lap time. Recognizing that finding the Michael Schumacher of AI is not enough, a team of researchers from University of Wuerzburg now wants to find the Mad Max of AI. Their new competition is called "Demolition Derby" and the goal is to "wreck all opponent cars by crashing into them without getting wrecked yourself". For this, they use the open-source TORCS game and a custom AI interface, allowing all and any AI researchers and enthusiasts (including you!) to submit their best and most aggressive controllers."

Submission + - Should computer games adapt to the way you play? ( 1

jtogel writes: "Many games use "rubberbanding" to adapt to your skill level, making the game harder if you're a better player and vice versa. Just think of Mario Kart and the blatantly obvious ways it punishes you if you drive too well by giving the people who are hopelessly behind super-weapons to smack you with. It's also very common to just increase the skill of the NPCs as you get better — see e.g. Oblivion. In my research group, we are working on slightly more sophisticated ways to adapt the game to you, including generating new level elements based on your playing style.

Now, the question is: is this a good thing at all? Some people would claim that adapting the game to you just rewards mediocrity (you don't get rewarded for playing well). Others would say that it restricts the freedom of expression of the game designer. But still, game players have very different skill levels and skill sets when they come to a game, and we would like to cater to them all. And if you don't see playing skill as one-dimensional, maybe it's possible to do meaningful adaptation? What sort of game adaptation would you like to see? I'm very interested in your feedback here..."

Slashdot Top Deals

Man is an animal that makes bargains: no other animal does this-- no dog exchanges bones with another. -- Adam Smith