In the past, when someone wanted to play online poker, they had to find a group of like-minded individuals who were on a similar search. Today the process is much easier thanks to online casinos and even more so since AI came on the scene.
Photo by Michał Parzuchowski at Unsplash
Artificial intelligence.
When machines receive a lot of data, they can look for patterns in that data to learn how to solve problems. The difficulty is that they are based on mathematical answers to problems. Think of the game of chess. Chess is a very conclusive game. Aside from the few initial moves, there is always the best play in chess, which is why chess AIs can annihilate even the best great masters. In the case of chess, the machines will evaluate the results of each move and select the one most likely to win.
Photo by Michał Parzuchowski at Unsplash
Why is poker different?
There is a psychological factor in Texas Hold’Em (although there is still a mathematical part), which is difficult for robots to master, but the addition of the lantern complicates matters. In chess, the best move can be determined based on the current actions and likely moves of the opponent, but Texas Hold’Em requires the use of intuition (even if logic says no). Players can pounce, overlay, or use other tactics that seem to defy logic.
Each player plays differently
So how do you modify the AI to suit each player’s strategy? Because knowing how your competitor plays is essential to becoming a solid poker player. Some players are “tight”, meaning they will only play one hand if their original two-card hand is statistically above average, so they will only score if they have a powerful hand. Other players are “loose” players who will play any hand and often only shout with good hands. The goal is to divide the game into smaller sections and tailor your approach as the game progresses. Thus, AI can use machine learning to identify and exploit holes in your opponent’s tactics, and now even win games.
There is also the problem of not knowing your opponent’s hand. In chess, both players are always aware of the exact state of the board. Because there are two hidden cards in poker, no player ever knows the actual condition. This makes it difficult to predict the final conclusion of the game. It’s also hard to count luck because poker is really a game of chance. You can start with the most amazing hand (a pair of Aces), but it would be shaky if all four of the five community cards were all on the same suit, as one color and the game could be in the hands of another. player.
What have the researchers done?
While there is no particular technique to explain this (artificial intelligence playing poker will always be an approximation), scholars try to solve the problem by developing a game and an abstraction where comparable hands are grouped. This makes it much easier for AI to assess the wide variety of possible hands that other players may have.
The current state of poker
Artificial intelligence is already capable of defeating professional poker players. In terms of strategy, it is also becoming more and more common to consult with artificial intelligence. More people than ever can use artificial intelligence to improve their methods. While traditional poker players learn by losing money, modern players learn by playing against machines. This is because scientists have created artificial intelligence to help professional poker players learn more about the optimal poker strategy, enabling them to become better players and earn more money in high-stakes games. bet.
This has had a significant influence on the poker industry. People are now more likely to try risky strategies and come up with risks that they would otherwise avoid because of the fear of failure.
Do people use AI to improve poker?
A professional poker player, Seth Davis, was competing in the World Series of Poker when he concluded a hand with no valued cards and few alternatives. Davis opted for Lighthouse and claimed he was “all-in” for the round, putting his $ 250,000 tournament registration fee at stake (not to mention his dignity).
Davis was relieved when his opponent folded and won his hand. That night, he evaluated his hand with PioSOLVER, an AI-based technology that has fundamentally changed the way poker is played in recent years. After a few seconds of entering the facts of the hand, the computer informed Davis that he had played his hand quite accurately and that his beacon was the correct movement.
Thanks to AI, this game based on luck is being transformed into a more calculated game with a clear strategy and method of improvement. What does this long-term improvement in AI mean? Given that a machine can outnumber people in a game of chance and uncertainty, how long will it be before AI makes accurate decisions that are bigger than a handful of cards?
