Computer AI has long been defeating humans in the game of chess, and earlier this year Google’s AlphaGo became the first to beat a world champion in the ancient Chinese board game Go, but now it’s time to settle the score in a game that really matters: a classic FPS. That’s what a group of AI researchers hope to see, with a new challenge inviting computers to face-off in Doom, but playing as humans would.
The “Visual AI Doom Competition” will be hosted by the 2016 Computational Intelligence and Games (CIG) Conference later this year, and now the group is accepting applications to find the best bots that can play a round of deathmatch against each other using the same learning techniques as a human.
But this isn’t the same kind of AI used for enemies when you play a game against bots. No, this kind of computer AI will not have all-encompassing knowledge of the game and its workings, instead it will have to rely only on what it “sees” as input. In other words, it will have to learn how to play the game, including things like the map layout, weapons, and opponent movements, in a similar way as a human does.
The bots will face-off in two challenges, with the first taking place on a Doom map that the programmers know in advance, with rocket launchers being the only weapon, and health being the only pickups. The second will be on a random map, with all weapons and items available across the environment.
The event is encouraging participants to use “reinforcement deep learning,” which is a method of AI that combines deep learning and reinforcement learning, or teaching through patterns found in large amounts of data, and teaching via rewards for successful actions. This is the same method that was used to train AlphaGo.
Bots must be written in C++, Python, or Java, and an early deadline of May 31st is set for warm-up matches, while final submissions must be made by August 15th. The best AIs will be selected to play in the tournament finals in September at CIG in Greece.
SOURCE Visual AI Doom Competition