AI program wins another poker tournament against human players

0

An AI system called Lengpudashi just beat a team of six poker players in a five-day competition in China.

The AI system won the tournament and $290,000 (£230,000).

This is the second time this year that an AI program has managed to defeat human poker players in a high profile series of exhibition matches.

In January, an AI program developed by computer scientists at Carnegie Mellon University managed to beat four human players in poker during a marathon match that lasted 20 days.

The AI program, called Libratus, demonstrated the ability to use strategy and reasoning as it won more than $1.5m (£1.2m) worth of chips from the human players.

The event, held at Rivers Casino in Pittsburgh, was live-streamed on Twitch.

Tuomas Sandholm, professor of computer science at Carnegie Mellon, called the event “historic”.

He said: “Heads-up no limit Texas hold ’em is in a way the last frontier of all the games,”

“Othello, Chess, Go, Jeopardy have all been conquered, but this remained elusive: this is a landmark in AI game-play.”

Jimmy Chou, one of the human players who participated in the poker event, was quoted by the BBC as saying: “The bot gets better and better every day. It’s like a tougher version of us.”

“The first couple of days, we had high hopes,” he said.

“But every time we find a weakness, it learns from us and the weakness disappears the next day.”

Jason Les, another participant, called the experience “incredibly challenging”. “We tried everything we could, but it was just too strong. It became very demoralizing,” he said.

Whilst the AI agent was designed to play Poker, the creators of Libratus believe that there are other possible applications – in a wide range of industries.

Prof Sandholm said:

“This is not just about poker.

“The algorithms can take information and output a strategy in a range of scenarios, including negotiations, finance, medical treatment and cybersecurity.”

“Now we have proven the ability of AI to do strategy and reasoning, there are many potential applications in future.”

 

You might also like More from author