What is the NZT brain?
The brain of the NZT consists of three main parts:
The NZT "brain" is a set of neural network decision-making algorithms, as well as databases on hands and players. Professional poker players are engaged in neural network training, and therefore our brain can find opportunities to exploit an opponent even where the human doesn’t see them.
A neural network is a program code that can be called Artificial Intelligence. The input is data that is processed according to certain principles.
These principles are kept secret by our company for security reasons. We call these principles the Black box. It is the Black Box that gives result in the form of a hint.
The more input data the neural network had, the better it would learn.
At the output, we get probabilities for each of the possible future actions of the opponent.
The task of the network: having received information in the form of Opponent Info + Action in current hand, to say with what probability the opponent will do this or another action.
● The network is constantly being trained using a database that includes more than 100,000,000 hands. The database is constantly being updated with the replacement of outdated and irrelevant information. This allows the neural network to always be aware of modern strategies and adjust its game in accordance with the current trends of the field, exploiting the weaknesses of opponents as effectively as possible.
● Opponent info is collected based on Hand History and includes VPIP, PFR and another 253 parameters that the neural network takes into account during the game.
We use a multi-layered neural network. This means that we give out parameters in some portions without mixing them, thereby obtaining a very branched decision tree.
Thus, the neural network makes decisions based on the statistical indicators of opponents and the actions performed in the distribution. The prompter analyzes every hand played with the opponent, and, if the situation requires it, can fold even strong combinations. You can play with this opponent for the first time, but if another player with a hint has already played against him, then the robot will make decisions taking into account all his past experience of playing against this opponent. This is artificial intelligence.
The development and testing of the brain took more than 15 years of work. More than 100,000,000 hands have been played.
● More than 12,000 hours of poker experts and developers are spent every month to improve brain function
● The NZT brain technically uses a neural network that trains itself, which means it develops faster than any professional player.
When your bot sits down at the table, most likely, he meets players with whom he has already played. Artificial intelligence (AI) has saved information about the opponent's strategy and now you get an advantage.