Borislav Ivanov and Rapid Chess

[Egidijus Zeromskis]

Re: Re : Re: Re : Borislav Ivanov and Rapid Chess

One more case of a mobile use.
http://goldensands.bulchesssummer.eu/rounds.html

[Paul Bonham]

Re: Re : Re: Re : Borislav Ivanov and Rapid Chess

His games were analyzed by Kenneth Regan and there's a statistically significant proof that the games could not have been played by a human. It is not sure at 100%, but the evidence would be strong enough in court, just as a DNA test. So, we can say, with 99.99+% certainty, that he is cheating.

I am disappointed in you, Felix, up until now you had impressed me as one of the more common-sense people here. "Statistically significant proof" is a concept you want to stay far, far away from. Where do you draw the line? Several years ago, someone "statistically proved" that the IQs of asian people are superior to any other racial group. Do you think no one would ever go so far as to suggest only asians should be allowed to procreate because of their "statistically proven" greater intelligence, and other races should be allowed to die out?

What you and Kenneth Regan fail to take into account in your statistical proof concept is the ability of the human brain to rewire itself and learn to play like Houdini given enough exposure to Houdini. It is entirely possible AND VERY LIKELY that a human being CAN LEARN to play chess like a 3000+ ELO computer engine. If you doubt this, then please explain to me how a human infant can learn to speak whatever language they are surrounded by between birth and about 2 to 3 years old, and can further learn rules of grammar for that language, and later learn to read and write that language? Explain to me how even an adult can learn to play a musical instrument by ear, without being taught by others or from written material?

To learn to play like a computer takes nothing more than immersement: simply play a computer engine game after game, day after day. Having a near-photographic memory would be a key aide in this, not because every possible position is going to come up in all your games, but because positions are similar enough that you can learn the "complex rules of grammar" pertaining to computer chess play.

Our ability to learn may diminish, but it most certainly doesn't disappear. If Ivanov did nothing but play Houdini game after game after game, day after day, week after week, month after month, he could learn to play at Houdini strength and make one of the top 3 or 4 or 5 Houdini move choices WITHOUT EXCEPTION in every chess situation. Out of all the possibilties mentioned in this thread, this is the most likely by far.

If you doubt that, think about how we learn language. We don't memorize sentences. We learn the component words, some phrases, rules of grammar and we learn to put them together to make up our OWN sentences and paragraphs. Some of us learn it better than others and write Monty Python skits or Seinfeld episodes. Ivanov may simply be one of the best learners of computer chess engine tendencies ever.

I hope both you and Kenneth Regan back off from every trying again to "statistically prove" anything. Do you realize that you are telling every chess player: "Do not stop making stupid mistakes. Don't you dare get as strong as Houdini, or we'll expel you from chess!"

Ivanov may be just the first of many. We are just in the infancy of the ELO 3000+ computer chess engine age. The effects are still rippling through the chess world. The ultimate effect will be a FIDE rule change within 25 years that will prevent computer chess engines from being relevant. The sooner the better, because it is ridiculuous to have to constantly worry about cheaters and those who are learning to play like Houdini.

This could possibly involve something similar to what I talked about in my Option Chess variant (formerly known as Double Move Chess until I discovered that name was taken). In Option Chess, each player has 4, 5, or 6 tokens that allow him / her to replace a single move with a double move during a game, subject to restrictions. Computer engines would take decades, perhaps dozens of decades, to adapt because choosing WHEN to best use a limited number of options is an AI problem different in kind from brute force calculation. It will take yet-to-be-realized advances in neural net techniques (rewiring the computer brain) for engines to ever be able to dominate humans at Option Chess.

[Egidijus Zeromskis]

Re: Re : Re: Re : Borislav Ivanov and Rapid Chess

Ivanov may simply be one of the best learners of computer chess engine tendencies ever.

When you start to play with a high performance, you must have a super special switch to go back to your true level of 1900.

[Paul Bonham]

Re: Re : Re: Re : Borislav Ivanov and Rapid Chess

When you start to play with a high performance, you must have a super special switch to go back to your true level of 1900.

Maybe the "special switch" was drugs or alcohol. Or lack of sleep. Or lack of confidence. All kinds of things can impair mental performance.

By your using the term "true level" I assume you believe he is cheating despite lack of proof. Please don't ever accept jury duty.

[Felix Dumont]

Re : Re: Re : Re: Re : Borislav Ivanov and Rapid Chess

Paul,
As you probably guessed by my way of thinking, I am a scientist. I cannot be convinced without facts and numbers ( it's probably worse since one of my majors is maths). Ken proves beyond reasonable doubt that this player played on a 3000+ level for many games, with the exact same moves as Houdini. Is it possible that his brain got "rewired" and that he is suddenly an extremely strong player? Yes, but it's extremely unlikely. I spent a few months in a neuroscience lab that focused on brain connections and while my research wasn't on this, what I learned is that the brain is way more complicated than that.

It would not be scientific to believe he suddently became the stronger player in the world by playing Houdini. He has no record of having a phenomenal memory, and even if it was the case, it would highly surprise me.
So, we have a case where both statistics and rational thinking agree. For me it's enough.
If I can't trust statistics then I should quit my job and stop doing research.

[Egidijus Zeromskis]

Re: Re : Re: Re : Borislav Ivanov and Rapid Chess

Maybe the "special switch" was drugs or alcohol. Or lack of sleep. Or lack of confidence. All kinds of things can impair mental performance.

By your using the term "true level" I assume you believe he is cheating despite lack of proof. Please don't ever accept jury duty.

Paul, you simple need to play more chess and less elaborate about person's chess capabilities. A person must be totally drunk for a week/weekend that his strength (performance) went down over 500 points during a whole tournament. And at the same time the person does not start to perform 500 points higher in 3 months.

[Paul Bonham]

Re: Re : Re: Re : Borislav Ivanov and Rapid Chess

Paul, you simple need to play more chess and less elaborate about person's chess capabilities. A person must be totally drunk for a week/weekend that his strength (performance) went down over 500 points during a whole tournament. And at the same time the person does not start to perform 500 points higher in 3 months.

Egidijus, you simply need to drink more alcohol and elaborate less about a person's individual tournament results.

[Egidijus Zeromskis]

Re: Re : Re: Re : Borislav Ivanov and Rapid Chess

Egidijus, you simply need to drink more alcohol and elaborate less about a person's individual tournament results.

Thnx for the advice but I'll pass by it.

[Tom O'Donnell]

Re: Re : Re: Re : Borislav Ivanov and Rapid Chess

So some people can play like Houdini because they have been exposed to it for so long and have absorbed the patterns, even if most of the machine's moves are based on its phenomenal rate of calculation. Meanwhile, other people who have been playing chess all their lives, many of whom are professional players and have spent thousands of hours sitting across from opponents, cannot discern the pattern of cheating in the behaviour of a few opponents.

You do realize that almost always the people convicted of crimes weren't caught in the act when they were arrested, right?

I am disappointed in you, Felix, up until now you had impressed me as one of the more common-sense people here. "Statistically significant proof" is a concept you want to stay far, far away from. Where do you draw the line? Several years ago, someone "statistically proved" that the IQs of asian people are superior to any other racial group. Do you think no one would ever go so far as to suggest only asians should be allowed to procreate because of their "statistically proven" greater intelligence, and other races should be allowed to die out?

What you and Kenneth Regan fail to take into account in your statistical proof concept is the ability of the human brain to rewire itself and learn to play like Houdini given enough exposure to Houdini. It is entirely possible AND VERY LIKELY that a human being CAN LEARN to play chess like a 3000+ ELO computer engine. If you doubt this, then please explain to me how a human infant can learn to speak whatever language they are surrounded by between birth and about 2 to 3 years old, and can further learn rules of grammar for that language, and later learn to read and write that language? Explain to me how even an adult can learn to play a musical instrument by ear, without being taught by others or from written material?

To learn to play like a computer takes nothing more than immersement: simply play a computer engine game after game, day after day. Having a near-photographic memory would be a key aide in this, not because every possible position is going to come up in all your games, but because positions are similar enough that you can learn the "complex rules of grammar" pertaining to computer chess play.

Our ability to learn may diminish, but it most certainly doesn't disappear. If Ivanov did nothing but play Houdini game after game after game, day after day, week after week, month after month, he could learn to play at Houdini strength and make one of the top 3 or 4 or 5 Houdini move choices WITHOUT EXCEPTION in every chess situation. Out of all the possibilties mentioned in this thread, this is the most likely by far.

If you doubt that, think about how we learn language. We don't memorize sentences. We learn the component words, some phrases, rules of grammar and we learn to put them together to make up our OWN sentences and paragraphs. Some of us learn it better than others and write Monty Python skits or Seinfeld episodes. Ivanov may simply be one of the best learners of computer chess engine tendencies ever.

I hope both you and Kenneth Regan back off from every trying again to "statistically prove" anything. Do you realize that you are telling every chess player: "Do not stop making stupid mistakes. Don't you dare get as strong as Houdini, or we'll expel you from chess!"

Ivanov may be just the first of many. We are just in the infancy of the ELO 3000+ computer chess engine age. The effects are still rippling through the chess world. The ultimate effect will be a FIDE rule change within 25 years that will prevent computer chess engines from being relevant. The sooner the better, because it is ridiculuous to have to constantly worry about cheaters and those who are learning to play like Houdini.

This could possibly involve something similar to what I talked about in my Option Chess variant (formerly known as Double Move Chess until I discovered that name was taken). In Option Chess, each player has 4, 5, or 6 tokens that allow him / her to replace a single move with a double move during a game, subject to restrictions. Computer engines would take decades, perhaps dozens of decades, to adapt because choosing WHEN to best use a limited number of options is an AI problem different in kind from brute force calculation. It will take yet-to-be-realized advances in neural net techniques (rewiring the computer brain) for engines to ever be able to dominate humans at Option Chess.

[Nigel Hanrahan]

Re: Re : Re: Re : Borislav Ivanov and Rapid Chess

So some people can play like Houdini because they have been exposed to it for so long and have absorbed the patterns, even if most of the machine's moves are based on its phenomenal rate of calculation. Meanwhile, other people who have been playing chess all their lives, many of whom are professional players and have spent thousands of hours sitting across from opponents, cannot discern the pattern of cheating in the behaviour of a few opponents.

When this smug cheater is finally caught, some people will be washing the egg off their face for days. (Reaches over for some popcorn, sees that he has none, and makes a mental note to get some in time for the fireworks.)

[Paul Bonham]

Re: Re : Re: Re : Re: Re : Borislav Ivanov and Rapid Chess

Paul,
As you probably guessed by my way of thinking, I am a scientist. I cannot be convinced without facts and numbers ( it's probably worse since one of my majors is maths). Ken proves beyond reasonable doubt that this player played on a 3000+ level for many games, with the exact same moves as Houdini. Is it possible that his brain got "rewired" and that he is suddenly an extremely strong player? Yes, but it's extremely unlikely. I spent a few months in a neuroscience lab that focused on brain connections and while my research wasn't on this, what I learned is that the brain is way more complicated than that.

You say it's extremely unlikely... HOW unlikely? You're the statistics expert, tell us how unlikely, put some numbers behind it.

The brain is way more complicated than what? This sounds very interesting, I'd like to hear more. I mean, if you know something in this area, perhaps you should contact the committee that is looking into chess cheating. Tell them that the brain can't be rewired to learn to play like Houdini. Provide the same evidence you've provided here.

It would not be scientific to believe he suddently became the stronger player in the world by playing Houdini.

Wait, you say you're a scientist, right? So where is the science behind that statement? What you have just made is a totally subjective statement. It is your opinion and nothing else, unless you can back it up with evidence.

You can't do that, because Ivanov is a novelty. He may be (and I don't know this for sure, but he has hinted at this in interviews) the first person ever to try and play like Houdini by spending hours every day, every day of the week, months at a time, with little interruption, playing games against Houdini.

In order to prove your statement, others must research this theory by doing what Ivanov claims to be doing. Others, with chess rating about 2000 or so, must commit to months of doing nothing but playing Houdini, then play against GM and IM level competition. The more that do this, the better for scientific evidence.

Until that is done, your statement above is garbage. Not admissable in a court of law.

So, we have a case where both statistics and rational thinking agree. For me it's enough.
If I can't trust statistics then I should quit my job and stop doing research.

Ok, I'm open minded: but exactly what statistics? You are posting here that Kenneth Regan has done such-and-such, and you provide no links to actual published data, actual research. You just expect us all to just believe it? Is this how you do science?

[Egidijus Zeromskis]

Re: Re : Re: Re : Re: Re : Borislav Ivanov and Rapid Chess

Ok, I'm open minded: but exactly what statistics? You are posting here that Kenneth Regan has done such-and-such, and you provide no links to actual published data, actual research. You just expect us all to just believe it? Is this how you do science?

let me that for you

or a direct link
http://www.cse.buffalo.edu/~regan/ch...and-report.pdf

[Egidijus Zeromskis]

Re: Re : Re: Re : Re: Re : Borislav Ivanov and Rapid Chess

Some updates on the case: http://www.chessbg.com/news/open/id/857
(used a translator)

"Bulgarian Chess Federation deems the case with Borislav Ivanov closed after chess player does not appear in the headquarters of the Federation to establish his innocence, where yesterday, 19/06/2013, the Wednesday had to be subjected to "special tests technology (lie detector) to prove objectively with scientific means that chess player does not use any unfair means in the game."

Not sure what it means: "closed". imo, that B.Ivanov can not play chess for 4 month under the Bulgarian flag.

[Kerry Liles]

Re: Re : Re: Re : Re: Re : Borislav Ivanov and Rapid Chess

Some updates on the case: http://www.chessbg.com/news/open/id/857
(used a translator)



Not sure what it means: "closed". imo, that B.Ivanov can not play chess for 4 month under the Bulgarian flag.

That whole process seems like a bit of a kangaroo court: appear at such and such a time for some tests (eg: lie detector)...
Perhaps the Bulgarian Federation has the authority to impose that, but it seems without any process unless I am missing some
other preliminary information. In any case, I am also not all that convinced of the accuracy of lie detectors.

It would be nice if someone could produce reliable information about how exactly this guy is accomplishing the cheating. Until
then, we have a massive puzzle and only allegations of cheating. I agree the correlation with top engines (Houdini in particular)
is an amazing bit of information, but is it enough to be certain? The reports that he takes nearly consistent and low time to
make moves is also highly suspect of course (someone who was interested in covering up their deceit would surely deliberately
alter the cadence?). Having someone simply 'guess' what an engine would play (with some decent chess background or not)
doesn't seem like an obvious way to beating a GM on a regular basis.

My feeling is that someday we will find out what technique he used to receive moves relayed by an accomplice or accomplices
and we will finally put aside nonsense like he 'trained himself' to think like Houdini...

[Felix Dumont]

Re : Re: Re : Re: Re : Re: Re : Borislav Ivanov and Rapid Chess

let me that for you

or a direct link
http://www.cse.buffalo.edu/~regan/ch...and-report.pdf

Thanks. But it was even mentionned in the very first post...

Dr. Kenneth Regan: That is exactly the philosophical question posed in my cover letter to the Ivanov report in January. I would like the chess world to determine this, in consultation with experts in other fraud-detection areas. I believe that a commission should be empowered to recommend procedures that would operate for judgments within the chess world. The procedures and judgments must be consistent with applicable laws, but should not require the use of general laws to implement sanctions against players.

Also as my report says, I cannot imagine the statistical evidence in any case being stronger than for Zadar. As it comes to Ivanov’s performance at the “Old Capital” Open in Veliko Tarnovo, even when the games from the first two rounds are included, I get significant deviations in both the MM and AE tests, and an IPR of 3149 with 97.5% confidence above 2940.

A bit of information in FM Valeri Lilov's new video that I regard as most surprising and important is that Ivanov took about ten seconds for most of his moves, regardless of position. GM Kiril Georgiev had stated this for his game against Ivanov in the rapid chess tournament in Kustendil (Bulgaria), where this would be less surprising. If that is true, then it meets the standards of observational evidence of cheating that I had in mind when composing my "Parable of the Golfers" policy page. I regard (outside-evidence + 1,000-1 odds) as having higher-precedence than (statistical-only 1,000,000-1 odds) in human terms, and perhaps this will inform your own deliberations about how a commission should regard the evidence. If this observation is true, then I will give a straight Yes answer to your final question, in my own opinion of guidelines that the chess world still needs to establish for itself. As it comes to my results with strong deviations, they are only based on the Ivanov’s games from Veliko Tarnovo and not Kustendil, for which I have only the game versus GM Georgiev. Sincerely, Ken Regan.

Ken Regan is a member of the FIDE Anti-Cheating Commitee
that will put forward proposals to the FIDE General Assembly in Tromso.

In any case, I don't see why I should continue arguing. The guy is cheating, there's not a single doubt. I spend my days with a supercomputer. Last time I checked, I could not do calculations that required 46 TB of RAM by myself. Maybe I don't learn fast enough? No doubt I should hire Borislav Ivanov to do my calculations... After all, he learned how to calculate billion of positions per second by playing against houdini!