When it's made the brilliant play, for example, we are able to show everybody, the bridge community, what was in the bot's minds, if you excuse my expression," Fantun said. "We can show why the bots made such and such a decision. But it has an important feature known in the tech world as "explainability." This type of AI is, in many ways, less complex. Apple or iPod? Artificial intelligence hoodwinked by handwritten note.artificial intelligence: Can you tell which 2020 headlines are real? So, it's very different from what we have seen in other games," Fantun said. "Also, we have some small neural networks that help mimic the the opponent's behaviour. Instead of playing billions of games and learning to predict outcomes, the AI simply learned the rules of bridge and got better over time with practice. They take in a lot data, then use that to make accurate predictions.īut because so much of the "data" in bridge is hidden from the opponent, NukkAI designed NooK as a "white box" system. Most AIs that have beaten humans at games rely on what's called "black box" systems. "We still have a roadmap in front of us." How did it win? "So even in bridge, there are other things to be solved," Fantin said. What's more, unlike real bridge players, NooK didn't collaborate with a partner. The tournament didn't include betting, which is usually a key component of bridge. The difference between the scores of the human and the AI were averaged over each set.īy the end, NooK won 67, or 83 per cent, of the 80 sets, making it the victor.ĭespite the win, Fantun admits NooK hasn't mastered all elements of game. (nukk.ai)Įach of the eight human champions played 10 sets of 10 games, while Nook AI played 80 sets of 10 games, for a total of 800 consecutive deals each. Jean-Baptiste Fantun is the CEO of NukkAI, a French artificial intelligence company. The NooK AI, meanwhile, took on the exact same role as the human champions, playing the exact same cards against the exact same opponents. But in this case, each human champion played alone, controlling both sets of cards, against a pair of AI bots called Wbridge5, which weren't created by NukkAI. ![]() Normally, bridge is played between two pairs of partners, each with their own set of cards. The tournament used a modified version of bridge. "So it makes it a very difficult game for the computer." Not your grandfather's bridge game "That's something that makes bridge very close to … real life," Fantun said. That makes it hard for an algorithm to make predictions about its opponent's next move. ![]() There are no secrets.īut in bridge, your opponents can work together against you, and you can't see their cards. Secondly, it has an element of mystery.įor example, in chess or Go, the game is one-on-one, and both players can see everything on the board. ![]() Firstly, bridge is a very social game that can involve both collaboration and deceit. The reason computers excel at chess but struggle with bridge is twofold, says Fantun. "What we've seen represents a fundamentally important advance in the state of artificial intelligence systems," Stephen Muggleton, a professor of machine learning at Imperial College London, told the Guardian newspaper. While computers have long been able to beat humans at a variety of games - including chess, checkers and Go - Fantun says nobody has managed to create an AI that can defeat a bridge champion before now. The bridge champions had every right to be cocky.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |