Poker-AI.org Poker AI and Botting Discussion Forum 2015-12-11T17:10:22+00:00 http://poker-ai.org/phpbb/feed.php?f=24&t=2943 2015-12-11T17:10:22+00:00 2015-12-11T17:10:22+00:00 http://poker-ai.org/phpbb/viewtopic.php?t=2943&p=6880#p6880 <![CDATA[Re: Kuhn Poker - Confidence level / Interations]]> Statistics: Posted by top_bottom_top — Fri Dec 11, 2015 5:10 pm


]]>
2015-12-11T17:09:10+00:00 2015-12-11T17:09:10+00:00 http://poker-ai.org/phpbb/viewtopic.php?t=2943&p=6879#p6879 <![CDATA[Re: Kuhn Poker - Confidence level / Interations]]>
If I understand your question correctly, all you have to do is compare the average strategy you get from CFR with the known Nash strategy to see how your solution converges with # of iterations.

This paper outlines the family of NE strats in Kuhn Poker.

https://www.jair.org/media/3402/live-3402-5992-jair.pdf

Statistics: Posted by top_bottom_top — Fri Dec 11, 2015 5:09 pm


]]>
2015-10-09T10:15:33+00:00 2015-10-09T10:15:33+00:00 http://poker-ai.org/phpbb/viewtopic.php?t=2943&p=6833#p6833 <![CDATA[Kuhn Poker - Confidence level / Interations]]>
How many interations would I need to run to have a high confidence level on my CFRM Kuhn Poker?
How do I calculate the confidence level?

I am using this Java code:
http://cs.gettysburg.edu/~tneller/model ... index.html

thx

Statistics: Posted by ocara — Fri Oct 09, 2015 10:15 am


]]>