The real idea behind networks, new ideas for networks, dogma of the past, chess of the future, opportunity...

[Jon Mike]

Upon the release of the Alpha Zero paper, the world of AI and chess realized a much more efficient model through a "new and more efficient" approach, after nearly half a century of using the previous techniques.  However, once again mankind may be only breeding gnats for races, when instead, finding even new approaches would put us leaps and bounds ahead of where we are today!  What is the goal behind chess networks?  Is it efficiency?  Master strength of play? Applications in other fields?  Most of these readers know that we are searching for the maximum global gradient descent, which is the best moves within the best plan, the Nash Equilibrium and the MinMax.  

For the past month I have traveled to "specialized" individuals with degrees of various sorts to discuss with them some very important matters regarding these fields.  Yet after these discussions I have realized just how specialized these topics are!  From mathematician to scientist to master chess player, they have failed to understand even the basics for my discussion and its importance. 

So here I am on lczero forums hoping to find others who specialize in the combination of 4 fields, game theory, programming, SF analysis and chess.  If these fields are your passion, then YOU should keep reading.

If you truly understand the following, then your understanding is greater than the specialists I have spoken with:

1. Solving chess is very important!  If we could solve chess, we would have to use a different approach, if this could be found it would be revolutionary across a vast number of fields.  The method of approach for the solution to chess is invaluable! (The real end goal of A0/LC0)

2.  But its currently beyond our capabilities!  The current methods have failed and the new method is more efficient, but still it is clearly mathematically impossible to "solve chess", given any currently known human methods using or hardware.

From this perspective, I have spent 15 years attempting new methods to solve complex problems such as chess.  June of this year I finally was able to formulate the solution in a form which is easily applied across chess and other fields.  

I spent 3 weeks rigorously testing the formula and have written 18 equations for (near) infinite chess. Since that time I have contacted officials from the CMI (queen's problem), USCF (chess), DeepMind (ai) and various colleges (mathematics).  Strangely, no one was interested.  It was quite disturbing to realize the difficulty of finding individuals who would even care for such a solution, since everyone discounts it as impossible!

I would like to share my findings with someone in the US, only if they specialize in the following 4 fields: game theory, programming, SF analysis and chess.  If that is you, please contact me.

[Edward Panek]

If you believe you have "solved" chess the next step would be verification/validation of your theory. How would you apply your solution in a way that is falsifiable?

[ratko...@desktopmetal.com]

@Jon Mike -- Can you post link where one could read more about your approach?

[Ghost]

Eagerly awaiting Jesse's commentary on this :)

[Jon Mike]

The formula could be understood as a process of 3 steps, which I call the "scientific method" for chess, (not to be confused with the control/variable scientific method).  Each mathematical step can be done can be done in your head in under 1 minute!  The result of this process gives the user PVs and replies in order!!!  

I have tested the veracity through Lomonosov tablebases and against Stockfish 9 (only discrepancies seem to be in closely evaluated positions...)

1. 50 initial tests in current WC games (passed perfectly and found each time magnus didn't play best)
2. 50 further tests in WC games of the past (found the best moves, mad respect for tal and fischer, but against this method their moves were flawed)
3. Full game test against chess-db.com's top ranked personality, Fischer (2875).  Method nearly won by move 6! By move 16 it was forced win!!!
4. Full game test against chess-db.com's top ranked living personality, Carlsen (2842) Dominated the strong engine with ease!!

When testing it against Stockfish9 in multiPV mode, it accurately outputs all the top PVs in order and their replies as deep as desired!

Believe it or not I traveled to various cities to demonstrate this to individuals, yet no one knew what to do about it or they didn't realize what Stockfish even was...

I do have very active plans to unveil this, but I need to have contact with an intelligent passionate player who specializes in all those fields...

[Jon Mike]

 Btw, I am ready for all the name calling and slander.  It will be worth it if only one individual who is a specialist in these areas makes himself known to me!

[Edward Panek]

What is your goal in this endeavor? What would success look like?

[Jon Mike]

A little background...

I called the St.Louis Chess Center with the request to "demonstrate" the solution to chess.  That was the wrong wording, I guess, because I was told there would be "No Demonstration" and that they were not interested.  They fumbled not knowing what to do with me and pointed me towards various colleges....I met with various mathematics professors who were stuck in the days of DeepBlue and out of touch with any importance to such a find.  I met with various USCF officers and a well known master, only to find myself face to face with biases so strong they claimed they couldn't see the proof of the demonstration, even when I used it to correctly produce a non obvious set of perfect answers in their selected GM complex positions.  Here the mathematicians and masters failed to have eyes to see because they did not want to, for the difficult positions at hand had impossible odds of perfectly solving.  It has been slightly sad after giving 15 years to chess to find there is no place for new thought, especially if it disrupts the current systems...I could go on and on about different things I have tried, and I didn't really want to put it "out there" but i am desperate to speak with the right persons who understand the fields involved. 

 

[Jon Mike]

My main goal: To find an individual who is proficient in the fields mentioned, so that as little background fill in as possible is needed (game theory, math, programming, chess and stockfish), in hopes of discussing specific details over a 10-20 minute phone call.

[Chris Whittington]

detailed explanatory paper first else BS

[Edward Panek]

" discussing specific details"

In technical terms in research this is called publishing a paper. Why not find a journal, determine their requirements , and publish a paper directly? Why wait for other people?

[Graham Jones]

Well, getting a paper published in a journal can take ages. DeepMind's A0 paper is still just a preprint. A preprint of course can be published very quickly. It's better than just sticking a document on a web site because the preprint archives tie the author's identity to the document. You can prove that you had your amazing idea before others.

[Jon Mike]

Edward, 

"Discussing specific details" is more clearly understood as getting intelligent counsel on a number of sensitive factors not to be mentioned here.

[Edward Panek]

Depends on the subject matter. I work in medical research and the papers are published relatively quickly. If the details of this "idea" are unique it should be easy to get provisional patent. And if true, this idea might work in other fields as well.

[Alexander Lyashuk]

I recommend writing a paper and publishing it to arxiv.org.

That's a modern way to publish mathematical discoveries.

There are many knowledgeable people who will gladly review the paper, and you will have a proof that you are an original author.

E.g. that's a way a proof for Poincaré conjecture was originally published, that's the way AlphaZero preprint was published, and currently it's really the best way to proceed.

--
You received this message because you are subscribed to the Google Groups "LCZero" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lczero+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lczero/dc98d771-36fe-4045-9bac-4cf8dcfac2bc%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Edward Panek]

Cornell! I know a researcher there!

[Zekka]

It sounds like right now, your problem is that you can't find real experts who will look at your work, because real experts aren't convinced you actually solved chess?

My suggestion is that you use your equations to win a chess championship, or at least beat an existing chess bot. If you really solved chess, you can probably beat Stockfish. That would earn you a lot of attention. You wouldn't have to release your research or rigorously prove it correct, but people would believe you.

Good luck!

[Zekka]

Oh, actually, I should preempt this. It sounds like you demonstrated that it had evaluations close to Stockfish's, or that it solved some puzzles. That's good, but it's different from what I'm suggesting, which is to win chess games.

I'm also a cranky, self-aggrandizing type of person (in different areas than you) and I'm used to being dismissed by people who just won't listen long enough to understand. I think saying that literally everyone is against you is pretty extreme though. I would take it as a sign that I'm doing something wrong. A lot of revolutionary findings get accepted eventually, even though some get ignored for a while.

If you're like me, you were probably dismissed because people weren't convinced duplicating Stockfish was that important, or that you were right when you brushed over minor differences in evaluation as unimportant.

I posted in a thread on a Go forum about this a while ago, where someone had come up with an algorithm to evaluate Go positions and kept excusing the algorithm's answers as partially correct, equally valid, etc. It was actually pretty clever and I thought the math involved was pretty. I'm bad at go, but I think in the canned examples he came up with he was good at showing that the algorithm knew if a group was going to die or not.

When it came to testing it, their algorithm didn't win games. It generated about twenty moves and the pro's move was usually one of the twenty -- on the boards the author presented. The author didn't do any stats to figure out how often this worked, and when the author tried to trim down the number from twenty, people usually found a position where the algorithm played badly. It would have been much easier if the author just published code and the algorithm started winning games.

[Jesse Jordache]

This is basically just the scientific method as I was taught in school.  I admire your rigor, sir.

[Zekka]

If his tests are rigorous as described and he got good results then I absolutely want to find out more! (But I still think using his solution to win games is the best way to present it.)

Even if what he has isn't a solution, it's potentially repurpose-able as a source for rules of thumb like the ones in Stockfish, especially since it sounds like he doesn't have to look at many board states to implement them. My guess is that if this is like the Go solver I saw, his rules are basically correct, but in edge cases no naive implementation can easily be made to work.

The Go guy's attempts to make his solution more specific ended up with him introducing rules like "if a group can be killed in N moves" which he believed was possible to implement cheaply but he failed to realize it was really complicated in edge cases, and his model fell apart when he tried to handle them.

I'm pretty woefully unqualified to evaluate what he did because my engine experience is all technical stuff, no conceptual stuff -- board representation, implementing MCTS.

[tun nay]

I highly doubt chess can be solved with current technology!

I once calculated the estimated accuracy in stockfish  in one game. If we assume the accuracy of stockfish as 99% in every single move (extremely optimistic) for ten critical positions in one game, the overall accuracy become 90% only. (which is quite far from perfection/solving chess). And if we assume 95% accuracy for 10 critical moves, then there is 60% accuracy for the whole game. 

So why do I think there can be 10 critical moves?

Most critical moves are between move 11 to move 30, or move 11 to move 50 depending on the game.(Cos some opening positions are almost bullet proof and many late middle positions can quickly end in TB access). Some games will have 4-5 critical moves only in case there are heavy pieces exchange, but some games will have more than 15.

Average game length of human games=40, Average game length of computer games=60

[Mike Kent]

Can you share some of these games? Especially wins against SF9?

[tun nay]

I highly doubt, we would see a win/lose game in solved chess, once chess is solved, there would be mixed ugly moves and beautiful moves.

What we would see in solved chess or 32 men Tb will behave the same as 6-7 men TB moves whereas some of  those moves can even be monkey moves. For example, KRB v KR endgame, player A, may even drop free rook, cos player A know that KB v KR is still a draw. Similar story, white player may start any move in initial position  and even dropping a pawn without reason cos white player knows than 15 men vs 16 men can still be a draw!!

[Jon Mike]

Yes, I can and will shared the games.  I am still waiting on a knowledgeable individual to make their selves known to me.  If this individual is passionate about theses fields, what's keeping them waiting?  The games I have mentioned will be posted but not right away.  The first game I call the awakening game because only viewing it would change how we understand chess to a greater degree of impact as the AZ games.  As such, these mentioned games will be released, but not yet and probably won't be public for another 3 months or so..the most common advice I get is to beat some local player (usually rated under 2000) or to "get my rating up" then people will listen.  WhIle the advice isn't wrong, it is clearly missing the point by such a great distance it short circuits my brain.  If I had a solution which could hold its own to sf, and destroy all other players, which I do, surely there is a more efficient route than 3 years or traveling to tournaments to get a Gm rating.  After all, I have defeated every super Gm engine I have tested, and don't have the time to play every local (and weak) local hero player.  In my state I do not get the chance for rated play except a few times a year.  I have to travel continously away from home in order to do this, and it has little to do with the solution and it's best method for presentation and testing.  Of course this is from my perspectime where the solution has been proven over and over in all tests, against the hardest puzzles and games.  

So what can't I solve with it?  So far only the mate in 50 to 500 moves, since I carry out the process in my head.  However, on paper it can be done, but such a process would take me days, and the rules of chess would ensure it was a draw anyway.  The solution works with or without the bounds of current boards size and draws due to too many moves without capture.  

This is not a known method, but was searched for by morphy, casablanca, tal and Fischer.  Alas, they  never actually found it as seen in their erroneous play (I tested their games to confirm).....

If you are an expert in the involved fields, hit me up...

[tun nay]

Sound like ARB troll, Good Luck in your career. :D

[Chris Whittington]

Yes, I can and will show the gold I made out of lead.  I am still waiting on a knowledgeable individual to make their selves known to me.  If this individual is passionate about theses fields of Chemistry and Alchemy and Atomic Physics, what's keeping them waiting?  The gold I have mentioned will be shown but not right away.  The first gold coin I call the Awakening Coin because only viewing it would change how we understand Physics to a greater degree of impact as the Hadron Particle Collider.  As such, these mentioned gold coins will be released, but not yet and probably won't be public for another 3 months or so..the most common advice I get is to buy something fairly expensive (usually rated under $2000) or to "get my credit-rating up" then people will listen.  WhIle the advice isn't wrong, it is clearly missing the point by such a great distance it short circuits my brain.  If I had a solution which could hold its own to Nuclear Fusion within the Sun, and destroy all other planets, which I do, surely there is a more efficient route than 3 years or traveling to conferences to get a Peer Review or two.  After all, I have made gold more pure than anyone have tested, and don't have the time to show it to every local (and weak) local hero person.  In my state I do not get the chance for escape from the asylum except a few times a year.  I have to travel continously away from home in order to do this, and it has little to do with the solution and it's best method for presentation and testing.  Of course this is from my perspectime where the solution has been proven over and over in all tests, against the hardest Reiman's hypothesis and Poincare Projections.  

So what can't I spend with it?  So far only the superyacht in 50 to 500 millions dollars, since I carry out the process in my head.  However, on paper it can be done, but such a process would take me days, and the rules of Physics would ensure it was a no go anyway.  The solution works with or without the bounds of current Large Hadron Accelerator size and fails due to too many moves without collisions  

This is not a known method, but was searched for by Einstein, Nostrodamus, Oppenheimer and Fischer.  Alas, they  never actually found it as seen in their ultimate poverty (I tested their bank balances to confirm).....

If you are an expert in the involved fields, hit me up...

[Jon Mike]

@tun nay,

@all, 

A.R.B is a chess system which involves making strange pawn moves in the opening followed by "perfect" play against SF.  Unfortunately, this system gives no definite ways to play the actual game.  There are no formula's and definitely not created from the strong fundamental logic found in my "scientific method".  The only thing which makes heads turn in the A.R.B system is the wins versus stockfish.  Yet, most believe as I do, that these games were fabricated.  

The A.R.B system fails in its delivery of communicating the "how" of the system.  With that being said, it is interesting how SF is weak in handling the A.R.B in the games which were posted online.  The ARB largely leaves the player in the dark and does nothing to explain how those perfect moves were found.  My conclusion was just that: beautiful and original A.R.B games, yet I believe it was human+stockfish vs stockfish, since no explanation was ever given as to how the moves were found.

The secret of chess, by Lyudmil Tsvetkov is another book explaining a system which is absolutely inapplicable for using during a chess match.  I have read it and it falls very short of useful.  

The scientific method for chess, however shouldn't be scoffed at.  I know its veracity by its fruit which is a direct result of the great laborers of the past...  

It was Fischer who stated he thought chess as a science, and himself a scientist and .... should be treated as such.  

It was Lasker, the mathematician who searched for the mathematical formula for chess.  

It was Capa, the logician who laid the fundamental logical framework to reduce complexity into simplicity.  

It was Tal which secretly but incompletely, found some of the real math involved and greatly enjoyed toying with his unique, yet primary understanding.

It was Turing who wrote the first algorithm and first modern computer as well as the "turing test" still currently the most efficient self improvement method.

It was Nash who laid the foundations in Game Theory, which is attempted to be applied in chess using "Minmax"  

It was CMI (clay mathematics institute) who offers 1 million dollars to find a different and much more efficient approach to solving the queen's problem.

It was Dennis Hassibas who shook the foundations of Chess & Machine Learning with Alpha Zero's different approach.

It was Leela Chess Zero which brought together those few individuals in the present, who understood the importance behind finding the global maximum gradient descent, equivalent to the Minmax, Nash or best move/best plan in chess.

There have been countless other individuals who have greatly contributed to mankind's effort in "solving chess".  The one thing they had in common was the foresight to see that such a solution had broad and powerful potential impact across a number of different fields.  

It was you and I who picked up where the laborers of the past had left off.  For whatever reason, (15 years of searching?), the mathematical solution landed upon me!?  I have clearly seen the greater need to release this to the world of chess rather than hold it selfishly for solely my benefit.  I promise it will be revealed, to you and to all, but not yet.  I am currently working on its proper delivery each and every day!  In the meantime, I would like to speak with an individual who is considered more knowledgeable than others, someone on LCZero who has high regard in the fields mentioned (3 out of 4 is sufficient).  I need a serious individual who is passionate about chess and has 10-20 minutes for a phone call, so that I may be able to cover greater details in a more private setting.

Tun Nay and others:  Do you have the ability to discern an honest individual?  Is your mind open enough to learn, or do you already know?  I am not a troll.  I am very similar to many individuals here.  I am logical.  I am skeptical.  I am a realist.  I love chess.  I love math.  I love game theory.  I love programming.  I love even you!   My plans are of mutual benefit for the whole of humanity and the chess world.  I will give all I have back to the whole, just as those greats in the past did!

I hope you are able to see my heart, hear my message and believe it is possible. 

P.s. I completely understand the inner dynamics of logic which cause the educated to dismiss me as a troll.  I am not upset, but fully expect such mis-perceptions.  All I can ask is don't put me in a stereotyped box, don't place me with ARB or Lyudmil (since I have made multiple public demonstrations and can explain every detail).  I am a different person, with a completely different approach, which is a combination of knowledge from the great masters of the past reduced down to a simple math process.

[Jon Mike]

@Chris,

As cute as your response was, (I actually laughed), I only feel sorrow in your response.  I wish that your eyes would be opened.

I actually work in those fields you mentioned, Atomic physics and Chemistry, and so you stirred up a passion inside me.  Did you know the individual who solved one of the seven million dollar millennium problems also recently solved a second one?  Did you know I invented the Atomic Force Microscope?  Did you know I have recorded the first videos of atoms?  Did you know I work in various quantum fields and have discovered new quantum states?  Did you know I broke the Ksp for various elements?  Did you know I successfully carried out the proposed experiments in Einstein's journal to create super-conductors and super-antennas?  Did you know the elements are specific vibrations or frequencies and only frequency and energy separate them?  Did you know all mass exists on the electromagnetic spectrum at the farthest highest energy regions?  No, so creating gold to you is hilariously bogus, but to a individual who understands his similitude to those great heroes of the past, to him it is made possible!

[Ghost]

should we send him the way of arb?

[Jon Mike]

IF someone did as I have claimed, it sure would be difficult to find anyone to take him seriously!  If he were to call upon even specialized individuals, all would discount him as insane.  Its really a tragedy.  I am offering and opportunity as I stated in the title!  What opportunity?  Individuals specializing in the aforementioned fields should contact me to find out!

[Robert Pope]

IF someone had done as you claimed, it would be very easy to take it seriously, because they would be providing evidence to support their position, not hyperbole.

[Jon Mike]

Hyperbole is an extreme exaggeration to get the point across.  I haven't exaggerated to any degree that I am aware.  I did do as claimed and I am willing to provide full evidence, but not here and not now.  This is merely an attempt to find the right individual with which to share in the rewards, but if that individual does not make themselves known, I am prepared to release it alone.

[Chris Whittington]

"share in the rewards"?

oh, so you are looking for an investment. How much do you want?

[Tony Asdourian]

My understanding is that there is a Nigerian prince who is eager to disburse some funds to whoever is willing to work with him just a little bit.  Perhaps you could team up with him and the two of you could reap your just rewards?

[tun nay]

I know ten ways to beat Stockfish and how to be  4000+ elo in chess. Since 3 months ago, I advertised those tricks in chess.com for US$1000  per trick but no one  believe me, and no one is interested in discussing with me :D
https://www.chess.com/forum/view/general/stockfish-is-very-weak-10-new-methods-to-beat-engines

[Jonathan Campbell]

I've often wondered at the solution to chess, as far away as this might seem to us now. I actually find it a little disconcerting (or maybe even depressing) that chess could be solved. It somehow detracts from the mystique of the game that I was never very good at but spent far too many hours playing (I'm old school and still playing on FICS some 20+ years after I first created my account). 

Musings aside, if you have solved chess, I'd like to know what the solution is. Can white force a win? Or as I expect, can black always manage a draw? I'd be blown away if somehow black can force a win... now that really would be something.

Incidentally, I'm new to this forum and really only aware of Leela through Kingscrusher's youtube channel (I highly recommended it). So while we are talking about solving chess, my guess is that Leela is converging (in a statistical sense) to a solution for chess. Does anyone have current statistics for the win/draw/loss ratio for black and white when the latest version of Leela plays herself?   

[Jon Mike]

@ Jonathan Campbell,

Hi Jonathan,

Thank you.  Although you are not a claimed specialist, you have a humble and teachable spirit.  Yes, the solution to chess does detract from the mystique.  Have you ever seen a Tal game?  Notice how Tals moves are full of "mystique"?  When a player uses the scientific method, they become free from the human misunderstandings and can enjoy the levels of GM and above.  There is freedom and justice built into the scientific method and it only unlocks the potential of the players.  Yes, I was concerned about this yet there is much room above 3000 elo.  But that strength of play is reserved for the future users of the scientific method.  

Yes, some find comfort believing the solution is so far away.  How many times have we heard the multitude saying, "it won't happen for thousands of years" only to find that we were wrong.  Regarding the initial state of the game, the system shows it is clearly a draw.  The system has even proven that 2 tempos from the initial position is also a draw mathematically, but only if 1...e6 is played.  Interestingly enough it has also proven that 3 tempos is a forced win for white (from starting position).  

Also noteworthy is an interesting form of zugzwang + zwischenzug, which I call, "Zwischenzeit".  This is a new chess principle, which is explained so that it is understood in the initial position.  Although I can prove this dynamic exists on the first move, I cannot prove that its ancestor (zugzwang) is or is not present in the initial position.  Most likely it is not, due to the flexibility of the minor and major pieces, but that discover is waiting in our not so distant future, at strengths 4000 elo and above.

When we understand that "drawn" games are not equal positions or equal journeys then maybe we will understand the math a little better.  In my system draws are distinguished by a number which reflects the strength of play, or the correctness of approach.  This insight alone would help many engines find their general gradient descent....

[Guy Apotheosis]

Jon - my background in three of the four areas may be sufficient to help you with your next steps.

1. Game Theory - I hold a PhD in mathematics and have studied Game Theory casually for most of my life.
2. Programming - I have worked as a Data Analyst and Software Engineer for multiple Fortune 500's and my dissertation included a large amount of custom code.
3. SF Analysis - none, though if your claims are as you suggest, this should not be a major obstacle.
4. Chess - I was a Class A player until I stopped playing tournament chess to pursue my academic studies.

If you are willing to share your work, I am willing to examine it -- but I begin from a position of being highly skeptical.  Extraordinary claims of a mathematical or scientific nature are commonplace on the Internet.  There is even a (semi-serious) Crackpot Index for such things and your posts in this thread are eerily similar to those you would find among those who claim to have a way to trisect an arbitrary angle, double the cube or square the circle.  Many people who submit such extraordinary claims end up shifting the burden ("No one has found an error in my proof!" or "The reviewers couldn't understand the depth of my ideas!").  While it may occur that I find a serious flaw in your work, that is neither the goal nor the standard.  My goal is simple: "Can I replicate your process and get the same results that you claim?"  If the answer is "No." then anyone with significantly less of a background than I have will likely struggle to get anything out of reading your work.  While it's possible that everyone else is simply incapable of understanding your thoughts on the matter, we must also recognize the possibility that you are not communicating properly.

With that in mind, there are several things that absolutely WILL NOT HAPPEN unless and until I am convinced that there is merit to your approach.
1. No phone call will take place; email will have to suffice. There are simply too many unsavory people online for me to deal with.
2. Nothing will be downloaded to my machine.  You will need to be able to explain your approach sufficiently in text alone that I can recreate it.
3. There will be no non-disclosure agreement or withholding of my findings.  You don't have to worry about anyone stealing your work because you will have the earliest emails on the subject (from when you first send your stuff to me).

If these terms are acceptable to you, let me know and we can move forward.

- Guy Apotheosis (not my real name)

[Guy Apotheosis]

It has now been over 15 months since the original post in this thread.  I offered my expertise to review Jon Mike's work but in that time I have received nothing to evaluate.  Unless Jon Mike jumps in, I think we can safely declare this dead and conclude there was never anything of actual value to consider.

- Guy Apotheosis (not my real name)

[Hadi Ghorbani]

Hi Jon

I amstrong chess player +2200 and also chesscoach  i am working to find best thinking system to find best move in chess i am working with evaluate of sf sources and  some principle in chess i am interesting some one do same thing in other side of the world i suggested you test this method on lichess make bot account and play with people and others bot i wanna hear more about your thinking process to solving chess! 

Best regards

[seff]

lmao, did you really expect something else?

[ER]

I argue that solving chess (game theoretic perspective) is not a worthwhile effort. Let’s consider a program that has a large opening book and a large endgame table. At any given position after the opening each move can either result in a sure win a sure loss or a sure draw. I argue that current top programs has a very low likelihood to select the lose for sure option. The vast majority is a draw option. Selecting the ones with the highest likelihood of a draw will lead to a draw against any opponent. If we explore this option taking it to its limit we would come up with an opening book that would result in drawn games. I think we all suspect that the game theoretic value of chess is 1/2, 1/2. What would we have gained by doing this?
If you agree with my perspective then we can state the goal more clearly. Play chess at the highest level with time constraint (minimal computational intensity) and minimal number of drawn games. In the process learning about intelligence in machines and people in a model (chess is a model AI case historically) , objective , and natural settings (the game was invented and played before the advent of computers) — that is a worthwhile goal

[Paul Berger]

On a related note, now that quantum supremacy is looming on the horizon. I'm hopeful it will be demonstrated for more and more generic classes of problems. 

AFAIK there were no convincing results to map game tree states to some form of quantum addressable (I mean inherently parallel) tree search or I am wrong?

Any pointer to relevant literature, papers or dedicated websites?

Best regards.

[Paul Berger]

Meanwhile, I just found a recent report, Quantum Computing: Progress and Prospects
Copyright National Academy of Sciences, 2019, 

available for free from here, https://www.nap.edu/download/25196

It contains this puzzling reference on page 64 sqq.

3.1.3 Grover’s Algorithm and Quantum Random Walks

While the QFT underlies many quantum algorithms, another class

of algorithms take advantage of a different method, called the “quantum

random walk.” This method is analogous to classical random walk meth-

ods, which probabilistically simulate progress in traversing some terrain.

Grover’s algorithm addresses the specific problem of finding the

unique inputs to a given function that will yield a certain output [14]. 3

Classically, this is a basic NP-hard search problem—that is, there are no

known polynomial time solutions. In the absence of information about

the nature of the function, the fastest known classical algorithm for this

problem is exhaustive search, or exploration of all possible inputs to

find the answer—a process that takes O(N) = O(2 n ) steps, where n is the

number of bits required to represent the input. Grover’s algorithm solves

this problem in O( N ) steps. While this is only a polynomial speedup

over the best classical approach, it could nonetheless be significant in

practice. As will be discussed in the next chapter, this could be sufficient

to compromise additional cryptographic operations. Moreover, this is

the optimal quantum algorithm for this problem, in light of the result of

Bennett et al. [15] showing that any quantum algorithm must take at least

N steps to solve this problem in the black box model.

The problem with the classical exhaustive search approach is that the

systematic testing of each possible answer is a blind guess-and-check:

each query provides no information about the answer until it is actu-

ally found. To get around this problem, Grover’s algorithm proceeds by

3 The problem can be more formally phrased as follows: If f is an efficiently computable

binary function that operates on strings of length n, find x such that f(x) = 1.

iterating a set of two operations on the qubits. The first effectively flags

the state corresponding to the correct answer by changing the sign of its

coefficient. The second, called the Grover diffusion operator, is then able

to slightly increase the magnitude of this coefficient. Together, the two

steps comprise a so-called Grover iteration, each application of which

increases the probability that the correct answer will be read-out upon

measurement. This procedure to increase the amplitude of the state(s)

containing the correct answer is an example of a general algorithmic

approach called amplitude amplification [16], which is useful in a number

of quantum algorithms.

The sequence of amplitude amplification operations can be viewed

as a sort of quantum random walk; however, Grover’s algorithm does

the “walk” backward, from a distributed state (analogous to all possible

endpoints of a random walk from a given starting point) back to a state

focused around the single correct component (analogous to the starting

point of the walk). A classical random walk approach can explore an area

proportional to the square root of the number of steps; the quantum ran-

dom walk can explore an area proportional to the number of steps. Hence,

the quantum algorithm provides quadratic speedup.

This technique is very versatile and has led to a number of quantum

algorithms providing polynomial speedups for specific computational

tasks. For example, there is a quantum walk-based algorithm for solving

the basic problem of determining whether the player making the first

move has a winning strategy in a combinatorial game (such as chess). The

naïve classical algorithm involves an exponential search of the possible

moves and outcomes, called the “game tree,” while the quantum algo-

rithm provides the quadratic speedup described above. More generally,

the quantum algorithm provides a quadratic speedup for evaluating any

AND-OR formula [17,18].

While Grover’s algorithm is often referred to as quantum “search,”

this is not really a valid application of the technique. To perform a true

quantum search, the set of data to be searched must first be represented

as a superposition of quantum states and for a quantum algorithm to

provide any speedup, this representation would need to be created in a

time much less than the number of data points, N—somewhere between

O(1) to O(logN). In the classical case, this data would simply be stored in

random access memory (RAM) and called when needed. However, while

RAM is a key element of classical computing, there is currently no robust

practical RAM equivalent that generates the needed quantum superposi-

tion state for a quantum computer.

It has been proposed that a quantum version of random access mem-

ory (RAM), or QRAM, could generate this data in O(log N) time [19],

although this has not been practically demonstrated. To achieve this, a

classical data storage unit would be supplemented with quantum logic

around the memory cells. A classical analogue to this structure, called

a content addressable memory, or CAM, exists, and solves this search

problem in O(log N) time. However, with CAM and with QRAM, getting

the data into the device in the first place still takes O(N) time, so either

approach will only be useful when multiple queries are performed on the

same data set—that is, the utility of CAM and QRAM, if it can be built,

grows in direct proportion to the number of times the input can be reused.

[smashu2]

Wait a minute why Sf 9 ?? you look foolish dude sf 99 we are at sf dev here wtf is this sf 9 ??
On Friday, July 27, 2018 at 2:55:02 PM UTC-4, Jon Mike wrote:

The formula could be understood as a process of 3 steps, which I call the "scientific method" for chess, (not to be confused with the control/variable scientific method).  Each mathematical step can be done can be done in your head in under 1 minute!  The result of this process gives the user PVs and replies in order!!!  

I have tested the veracity through Lomonosov tablebases and against Stockfish 9 (only discrepancies seem to be in closely evaluated positions...)

1. 50 initial tests in current WC games (passed perfectly and found each time magnus didn't play best)
2. 50 further tests in WC games of the past (found the best moves, mad respect for tal and fischer, but against this method their moves were flawed)
3. Full game test against chess-db.com's top ranked personality, Fischer (2875).  Method nearly won by move 6! By move 16 it was forced win!!!
4. Full game test against chess-db.com's top ranked living personality, Carlsen (2842) Dominated the strong engine with ease!!

When testing it against Stockfish9 in multiPV mode, it accurately outputs all the top PVs in order and their replies as deep as desired!

Believe it or not I traveled to various cities to demonstrate this to individuals, yet no one knew what to do about it or they didn't realize what Stockfish even was...

I do have very active plans to unveil this, but I need to have contact with an intelligent passionate player who specializes in all those fields...

[Adam Kirby]

I see Jon Mike has not seen fit to respond in over a year even though he still posts here.

Things that make you go hmmm.