Poker-AI.org Poker AI and Botting Discussion Forum 2013-07-10T16:01:54+00:00 http://poker-ai.org/phpbb/feed.php?f=24&t=2520 2013-07-10T16:01:54+00:00 2013-07-10T16:01:54+00:00 http://poker-ai.org/phpbb/viewtopic.php?t=2520&p=4411#p4411 <![CDATA[Re: Difference between CS-CRM and Vanilla]]>
Vanilla: Nothing is sampled. You traverse the whole tree, including all board card combinations.
Public chance sampling: Board cards (public information) are sampled. Obviously this is the same as vanilla if there is no public information.
Chance sampling: Board cards and hole cards (private information) are sampled.
External sampling: Chance sampling + opponent's strategy is also sampled.

etc... There is more to sample.

Statistics: Posted by amax — Wed Jul 10, 2013 4:01 pm


]]>
2013-07-10T15:37:00+00:00 2013-07-10T15:37:00+00:00 http://poker-ai.org/phpbb/viewtopic.php?t=2520&p=4409#p4409 <![CDATA[Re: Difference between CS-CRM and Vanilla]]> Hipp wrote:

Vanilla: every iteration we pick randomly only board cards. We traverse whole decisions tree and players possible hole cards.
Chance Sampled CRM: every iteration we pick randomly board cards & players hole cards, and traverse whole decisions tree.

Correct

counter-factual regret minimisation or counter-factual regret minimisation

Statistics: Posted by spears — Wed Jul 10, 2013 3:37 pm


]]>
2013-07-10T10:50:17+00:00 2013-07-10T10:50:17+00:00 http://poker-ai.org/phpbb/viewtopic.php?t=2520&p=4405#p4405 <![CDATA[Re: Difference between CS-CRM and Vanilla]]> http://poker-ai.org/archive/www.pokerai ... a&start=20

I'm sorry. The shortcut is "CFR" not "CRM" (I don't know why). But on forum you call it CFRM.

Statistics: Posted by Hipp — Wed Jul 10, 2013 10:50 am


]]>
2013-07-09T20:51:33+00:00 2013-07-09T20:51:33+00:00 http://poker-ai.org/phpbb/viewtopic.php?t=2520&p=4400#p4400 <![CDATA[Re: Difference between CS-CRM and Vanilla]]> Statistics: Posted by spears — Tue Jul 09, 2013 8:51 pm


]]>
2013-07-09T19:55:00+00:00 2013-07-09T19:55:00+00:00 http://poker-ai.org/phpbb/viewtopic.php?t=2520&p=4399#p4399 <![CDATA[Re: Difference between CS-CRM and Vanilla]]> Statistics: Posted by Hipp — Tue Jul 09, 2013 7:55 pm


]]>
2013-07-09T14:07:31+00:00 2013-07-09T14:07:31+00:00 http://poker-ai.org/phpbb/viewtopic.php?t=2520&p=4397#p4397 <![CDATA[Re: Difference between CS-CRM and Vanilla]]> Statistics: Posted by spears — Tue Jul 09, 2013 2:07 pm


]]>
2013-07-09T13:29:44+00:00 2013-07-09T13:29:44+00:00 http://poker-ai.org/phpbb/viewtopic.php?t=2520&p=4396#p4396 <![CDATA[Re: Difference between CS-CRM and Vanilla]]> ?

Statistics: Posted by Hipp — Tue Jul 09, 2013 1:29 pm


]]>
2013-07-05T22:04:32+00:00 2013-07-05T22:04:32+00:00 http://poker-ai.org/phpbb/viewtopic.php?t=2520&p=4391#p4391 <![CDATA[Difference between CS-CRM and Vanilla]]> Vanilla: every iteration we pick randomly only board cards. We traverse whole decisions tree and players possible hole cards.
Chance Sampled CRM: every iteration we pick randomly board cards & players hole cards, and traverse whole decisions tree.

Am I right? If not - can you please explain me the difference in a simple words?

Statistics: Posted by Hipp — Fri Jul 05, 2013 10:04 pm


]]>