~
~
Pluribus’s self-play produces a strategy for the entire game offline, which we refer to as the blueprint strategy. Then during actual play against opponents, Pluribus improves upon the blueprint strategy by searching for a better strategy in real time for the situations in which it finds itself during the game.
Only by then, on the actual live gameplay, do they adapt and adjust accordingly.
So it can't be said and is not to study the opponent's play by hours.I guess you're right. The reality is a bit different from what I imagined. I think I've been misled by a video I watched 2-3 years ago.
I still think that no AI can learn through self-play
only, how to beat human poker pros, that improvement of the the blueprint strategy during the game is very important, but apparently you are right, the AI doesn't need to study the opponent's play by hours.
Well for now things when it comes to AI, the training of robots is relatively slow, but compared to the learning that a human has, it is very fast, so for now Natural Intelligence has a lot to do with the fact that there are things that are above AI, of course I think that everything goes in the sense of Logic, because human logic has a lot to do with learning any AI, so that is where the great detail is of how to do so that they have a better path to learning, but things are going to move forward, when it comes to technology this is advancing at a great pace, something that can happen over time, for now things with AI are very slow, AI does not have as much recognition or awareness about some things,. It hasn't been measured against the best players in the poker world, but these AIs can already learn to analyze very quickly, and the demo or beta mode they are in now is something that is like a baby, who is just starting to crawl, But that baby grows fast and when we see him, he runs much faster than a human.
Poker players around the world know a lot due to their experience and also their intuition, and these things are difficult for an AI to learn, how to learn to intuit? How do you learn that some things are different and obey merely human logic? How to make a decision at the right time and make it the most appropriate, thinking about other things and not just theoretically? The AI is only interested in thinking to obtain profits, and win everything, but once you get there it is good that you know at least how the best players in the world think and there are things that cannot be done just like that at once, what I say is When an AI develops more in two or three years, I think that human beings will know that they have to be stopped in some way, we should not give it so much importance and not give it so many wings so that they do not get unpleasant surprises.