Monday, September 11, 2017

Carnegie Mellon Artificial Intelligence Beats Top Poker Pros

Tuomas Sandholm (center) and Ph.D. student Noam Brown developed Libratus.

Classmate Weixiang Wang has just given you a good reason to fold. As Carnegie Mellon's website details, Libratus, an AI system designed at CMU has just surpassed human competitors at the poker table. Here's a key insight into how the system was able to self-correct and improve:

“After play ended each day, a meta-algorithm analyzed what holes the pros had identified and exploited in Libratus’ strategy,” Sandholm said. “It then prioritized the holes and algorithmically patched the top three using the supercomputer each night. This is very different than how learning has been used in the past in poker. Typically researchers develop algorithms that try to exploit the opponent’s weaknesses. In contrast, here the daily improvement is about algorithmically fixing holes in our own strategy.”
Sandholm also said that Libratus’ end-game strategy, which was computed live with the Bridges computer for each hand, was a major advance.
“The end-game solver has a perfect analysis of the cards,” he said.

2 comments:

  1. This is fascinating. Metaculus, a community prediction platform for rating the likelihood of various technological and scientific outcomes, has also been tracking the progress of this AI milestone (https://www.metaculus.com/questions/345/will-a-poker-bot-beat-the-best-human-players-at-online-multi-player-no-limit-texas-holdem/). Aside from a few nuances, it's basically been resolved with Libratus' performance.

    ReplyDelete
  2. some on line casino provide unfastened money for sure games. And others offer to double or triple the deposits of the player. however, are on line Casino Bonus Catalog fantastic to online on line casino gamers?

    ReplyDelete