blonde poker forum
Welcome, Guest. Please login or register.
April 22, 2021, 12:09:12 PM

Login with username, password and session length
Search:     Advanced search
2259975 Posts in 66603 Topics by 16869 Members
Latest Member: HughOlssen
* Home Help Arcade Search Calendar Guidelines Login Register
+  blonde poker forum
|-+  Poker Forums
| |-+  The Rail
| | |-+  AI program beats humans in poker game
0 Members and 1 Guest are viewing this topic. « previous next »
Pages: [1] Go Down Print
Author Topic: AI program beats humans in poker game  (Read 2080 times)
RED-DOG
International Lover World Wide Playboy
Global Moderator
Hero Member
*****
Offline Offline

Posts: 45585



View Profile WWW
« on: February 01, 2017, 10:02:15 AM »



 http://www.bbc.co.uk/news/technology-38812530
Logged

How can we change the world?
One random act of kindness at at time.
TightEnd
Administrator
Hero Member
*****
Offline Offline

Posts: I am a geek!!



View Profile
« Reply #1 on: February 01, 2017, 11:21:43 AM »

and more

"Poker computer trounces humans in big step for AI

Libratus, an artificial intelligence robot, has won chips worth $1.5m from four of the world’s top poker players in a three-week challenge at a Pittsburgh casino"

https://www.theguardian.com/technology/2017/jan/30/libratus-poker-artificial-intelligence-professional-human-players-competition?CMP=twt_gu
Logged

My eyes are open wide
By the way,I made it through the day
I watch the world outside
By the way, I'm leaving out today
teddybloat
Sr. Member
****
Offline Offline

Posts: 726


View Profile
« Reply #2 on: February 02, 2017, 08:20:42 AM »

Interesting challenge for a couple of reasons.

Heads up poker is perilously close to being fully solved, a year or two maybe.

Liberatus wasn't attempting to solve poker, and  indeed the program was being tweaked between sessions to plug leaks. It isn't a bot in the sense of it having a set strat or being able to adjust on the fly without human intervention.

In the previous challenge the team behind liberatus even deployed different versions of the bot against each of its human opponents. I haven't followed the challenge closely enough to know if this is the case again this year, but there is no reason to suspect it to be any different.

The programmers aren't publishing any data on the exploitabily of liberatus, which is something that I believe is fairly easily calculated.
Logged
Longines
Gamesmaster
Hero Member
*****
Online Online

Posts: 3741


View Profile
« Reply #3 on: December 19, 2017, 02:47:33 PM »

The developers have published a paper in Science: http://science.sciencemag.org/content/early/2017/12/15/science.aao1733.full
Logged
Pages: [1] Go Up Print 
« previous next »
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Valid XHTML 1.0! Valid CSS!
Page created in 0.175 seconds with 21 queries.