Online gaming transformed by a player you'll never be able to beat: One robot has played more poker than all of humanity combined

 
Sarah Spickernell
Follow Sarah
Which hand? It doesn't matter - you'll lose anyway (Source: Getty)
You might as well give up, people – you'll never be the best poker player in the world.
There's now a computer programme so good at it, the Canadian scientists who developed it say it will never make a mistake.
They describe the new software as having “solved” the game, with an algorithm strategy so close to optimal, "it can't be beaten with statistical significance within a lifetime of human poker playing".
It was programmed to play 24trn simulated poker hands per second for two months, which means it has probably played more poker than all of humanity has ever experienced.
“We can go against the best (players) in the world and the humans are going to be the ones that lose money,” said Michael Bowling, lead author of the study.
“Poker has been a challenge problem for artificial intelligence going back over 40 years, and until now, heads-up limit Texas hold'em poker was unsolved.
“We define a game to be essentially solved if a lifetime of play is unable to statistically differentiate it from being solved at 95% confidence.”

Learning to regret

Throughout the relatively short history of artificial intelligence in online gaming, a number of cases have arisen where computer algorithms were better than the top human players.
But these winning machines were for "perfect-information games", where all players, human and non-human, are aware of everything that has happened in the game up to the point of having to make a decision.
In the case of poker, however, players do not know which cards have been dealt to other players, and this as been a stumbling block for developers in the online gaming industry.
The software has taught itself to overcome this by learning to “regret” and remembering every decision that does not lead to the best possible outcome.

Related articles