#41
|
|||
|
|||
Re: Phil Laak vs. the bot
[ QUOTE ]
Feelings? Lol. Poker players who play feelings consistently are losing players. [/ QUOTE ] Someone should get that memo to Jennifer Harman. |
#42
|
|||
|
|||
Re: Phil Laak vs. the bot
[ QUOTE ]
Chess is about as strategic of a game in existence. Chess is infinitely more complex than poker. [/ QUOTE ] Nothing more incorrect has ever been written in the entire history of mankind. (Except perhaps the Bible which does come in a very close 2nd) |
#43
|
|||
|
|||
Re: Phil Laak vs. the bot
[ QUOTE ]
Good chess computers are taught to look at millions of different possibilities for future moves precisely because you don't have all the information. If an opponent moves his knight to b5, you have absolutely no information on why he did it. You have to look at all the possibilites and make an educated guess. [/ QUOTE ] When programmers talk about "perfect information" they mean that they know the entirety of the game state. The motivation behind a move means absolutely nothing. If the computer is given a chess position, it would treat it the same way if it got there by playing through the game or if you had simply set it up on the board. The preceding moves do not figure in the calculation. |
#44
|
|||
|
|||
Re: Phil Laak vs. the bot
[ QUOTE ]
Bluffing and catching bluffs is just game theory. Sklansky covers this in his book. Chris Ferguson is a master of game theory. There is a correct percentage to bluff and to call bluffs. Game theory is can easily be programmed and it's a lot more effective in the long run than playing feelings. [/ QUOTE ] Do you even know how to solve a non-trivial game theory problem? We'd need around 1 billion BlueGene/L's working in parallel to get a decent answer to the full HU limit problem (button posts SB, 4 bet cap) within a year. Marv |
#45
|
|||
|
|||
Re: Phil Laak vs. the bot
It would be interesting to raise 63o in EP against a bot..then bet out on a AJ9 flop....guy did this to me on 3/6 table other day and farmed my ass since my play is predictable, similar to a bot I guess. (yes I had AQ) and folded on the turn...
|
#46
|
|||
|
|||
Re: Phil Laak vs. the bot
[ QUOTE ]
A perfectly programmed bot will not lose in the long run. [/ QUOTE ] Keyword---perfectly. The randomness of a player's actions simply cannot be accounted for by programming. What makes a player good? The ability to shift gears and adjust to opponents. How would a bot respond to a player shifting gears? Not to mention that the bot would have an unmeaningful sample size on nearly every player, how would it account for a stretch of 1000 hands where a player has a VP$IP of 12% and AF of 0.6 and then over the next 1000 hands switches to a VP$IP of 36% and AF of 3? |
#47
|
|||
|
|||
Re: Phil Laak vs. the bot
Imagine, trying to play a game of chess, where you couldn't see any of your opponents pieces [img]/images/graemlins/wink.gif[/img]
|
#48
|
|||
|
|||
Re: Phil Laak vs. the bot
Nothing, just send them to Guantanamo Bay...
The idea that law enforcement or politicians could be summoned to assist is funnier than anything I've read in weeks. |
#49
|
|||
|
|||
Re: Phil Laak vs. the bot
My experience in online video games such as diablo 2 is that they could tell people were using bots, especially when they had dedicated servers, it detects packets being sent and patterns and such. Patterns in betting timing and stuff can be detected by the poker sites and monitored as well. Video game multiplayer games have made great strides in preventing cheating and botting, and should give online casinos a head start. I hope and believe online casinos will be able to negate botting and cheaters effectively. Google macros for more info.
|
#50
|
|||
|
|||
Re: Phil Laak vs. the bot
[ QUOTE ]
[ QUOTE ] A perfectly programmed bot will not lose in the long run. [/ QUOTE ] Keyword---perfectly. The randomness of a player's actions simply cannot be accounted for by programming. What makes a player good? The ability to shift gears and adjust to opponents. How would a bot respond to a player shifting gears? Not to mention that the bot would have an unmeaningful sample size on nearly every player, how would it account for a stretch of 1000 hands where a player has a VP$IP of 12% and AF of 0.6 and then over the next 1000 hands switches to a VP$IP of 36% and AF of 3? [/ QUOTE ] A perfectly programmed bot with a HH database backing it up would be unbeatable, and there is nothing in online poker a bot couldnt be programmed to handle perfectly. |
|
|