Is it inevitable that poker bots will eventually be the majority?

[Gregg777]

[ QUOTE ]
I am confident bot detection can outpace bots means to pass the tests. Funky letters at first, then picture identicication, then logic problems, who knows what the future will be, but i'm sure humans are clever enough to devise means where bots wont pass the Turing Test.


[/ QUOTE ]

Bingo...

[iH8poker]

[ QUOTE ]
better question...when will bots overrun B&M tables? Or does this somehow correlate to when true AI emerges. I bet Johnny 5 can kick some poker ass!

[/ QUOTE ]

I have been thinking about this a lot lately...what would it take and WHO HAS A BETTER poker face than a non-feeling machine? The bot could even wink to throw ppl off...or a person from India (or they sound like they are from India, but they are really one of those hairy Russians) could sound like they are speaking from the bot.

I think Johnny 5 actually played poker in one of his movies or played the roll of a woman, I can't remember. Eitherway, I think in the last movie he had lasers or something. The bots got way more advanced during those movies than they have in real life (at that time, thanks to Hollywoods *magic*).

I wish they still worked with Johnny's programmers to make him learn to sing Help Me Rhonda (or play it over a phone, that was classic).

[david050173]

[ QUOTE ]
[ QUOTE ]
Now my bot just needs a hardware component (someway to press keys and read the screen. Neither is super hard). If I can make 1 million dollars a year by running my bot I can afford to spend a lot of money on hardware.

[/ QUOTE ]

If I wanted, I could empty your bank account just starting out from knowing your username on here, you can't stop crime or the potential of crime, but if you make it hard few people find it worthwhile.

[/ QUOTE ]

I could also found a company, take it public, and make billions so why would anyone play poker? It is trivial to do the hardware (ie all the info is publically available on the web and I have seen a half dozen undergraduate projects that do both just among the people I interview for jobs). You hacking of the bank account is theoritically possible but far more unlikely.

If the reward of a bot is small, they might not become common place. If the reward is large, they are invetiable.

[mjkidd]

[ QUOTE ]
There is a famous quote in computers/computing, some famous comptuer guy (apparently not famous enough for me to remember his name) basically said

"I think we are beginning to reach the limits of what computers can do. Although one should be careful with such statements as they can look very foolish in several years."

and this quote was from like the 1950's or something.

[/ QUOTE ]

The really funny thing is that this statement is sort of true. The computer I'm typing this on isn't any "smarter" than the computer that this famous computer guy was talking about. It's just much, much, much faster and much, much smaller.

Outside of chess, no one has even come close to building an AI that beats a complex game, let alone a game with incomplete information. I'd be surprised if any bot designed in the next ten years could beat a limit holdem game full of experts, or have a higher win rate than an expert in any given game.

[PLOlover]

[ QUOTE ]
Outside of chess

[/ QUOTE ]

backgammon

[Arnfinn Madsen]

But the current chess and backgammon bots play far from perfect, they use a lot of shortcuts/evaluations to determine the best play, they do not actually calculate it, it is sufficient though to beat humans. In backgammon there are players that can play better vs opponent x than the bot can (mostly due to realizing how the opponent will handle cube and using sub-optimal plays to incite mistakes), but they still lose to the bot. The same will be the situation in poker I guess, we are far away from the bot beating the best players and much much farther away from the bot playing better vs humans than the best humans can do.

[Sniper]

Arnfinn, you should read the Party Bot thread... there have been several 2+2ers who have noted that the Bots that have been identified, were beating them at 10/20 - 30/60.

[Arnfinn Madsen]

Yeah, I read, it happened earlier than I thought, mainly because I think it is very few people who are good enough to give the input, I stand by my statement that the player giving the input must be a much better player than the bot, I just react when people sort of think as the bots as perfect, optimalization in poker involves an understanding of human psychology and understanding human psychology is still the hardest part for computers. We don't understand much of it ourselves yet, so no way we can make a computer that understands it (at that point btw 99% of all customer service personell will be fired so Full Rake is right that the poker implications are kind of insignificant).

[mjkidd]

[ QUOTE ]
[ QUOTE ]
Outside of chess

[/ QUOTE ]

backgammon

[/ QUOTE ]

Fair enough. But backgammon is an open-information game, and it's heads-up.

[maurile]

[ QUOTE ]
How do you program intuition? All great players have this to some degree.Every good player knows the math and numbers,but there is a feel for things humans have.Some have more or less of this,but it cannot be taught or programmed,and to deny that this exists is false.

[/ QUOTE ]
What do you think human intuition is -- magic? It's the processing of information by the brain. Computers can do this. (It's 100% certain that computers can do this, because your brain can do it, and your brain is a computer.)