Will computers ever solve chess?

; )

Game Theory can be useful for understanding the nature of games and strategy in general, but that is not its primary application by a long shot.

Game theory is "the study of mathematical models of conflict and cooperation between intelligent rational decision-makers". Game theory is mainly used in economics, political science, and psychology, as well as logic, computer science and biology.^[1] Originally, it addressed zero-sum games, in which one person's gains result in losses for the other participants. Today, game theory applies to a wide range of behavioral relations, and is now an umbrella term for the science of logical decision making in humans, animals, and computers.

Did you peruse the MIT paper I linked?  It's a misnomer to call it Game Theory, and much like Gamification, which is another subject that has "game" in its title that has nothing directly to do with a game like chess, there are metric tons of people taking college courses on these subjects who are quite surprised to learn that it won't really help them play Chess, or CounterStrike, or Football...when I took Wharton's Gamification course, 50% dropped the class after the first 2 weeks because didn't understand that Gamification is about marketing and reward systems, not gaming.  Game Theory is about decision making, and it all started with zero sum games, because they are simple to model, but is more about decision-making among groups of interactive "players".  On that score, I will grant you without reservation that Game Theory is definitely more useful for Poker than for Chess.

Thanks for finally admitting that you did the flawed analysis and stand behind it .  We're making progress.  Now you just have to realize and admit you're wrong.

I do not believe a computer will solve chess in our lifetime, nor our children's lifetime, nor the lifetime of our children's children.

The number of estimated moves in a chess game is 1.0E+120

If every single person on earth, began playing now, a game a nanosecond, in a trillion years we would have played 2.20752e+38 games. How close would we be to having played them all?

2.20752e-83

We haven't even come close at that point. Using a 24 hour day as a representation, we'd still be on the first nanosecond of that 24 hour day. There isn't a man alive who could even conceive of such a machine as it would take to solve such a thing, if ever it could be solved at all.

Wow, sad that people would spend money in colluage corse they entered in blind!

Regarding game theory, it's worth pointing out that it is only two player zero sum games that have the sorts of meaningful results that our intuition desires.

A very simple result is that is any n-player game is equivalent to an (n+1)-player zero sum game. This is why two player games such as "the prisoner's dilemma" and "chicken" are so strange (they are equivalent to 3 player zero sum games, where the third player takes no active part but gets a reward so everything balances. In the game of the prisoners' dilemma, the third player can be considered to be justice, in the case of chicken, it can be viewed as death. 

With this view of such games as multiplayer zero sum games, it can be seen that the logical self-serving strategies that lead to bad results for all participants are really collaborations with the third imaginary player (snitching in the first game, being suicidally reckless in the second). From this it can be seen that good strategies are those where the human players collaborate rather than competing, so as to defeat the third player.

This principle happens to be true in extremely important applications of game theory concerning the policies of individual countries in a shared world: simplistically self-serving ones "I just want to get the best deal for my country" can be bad for all countries, as well as in many other human contexts.

I hope you don't mind the diversion into a fascinating topic.

We here at "Flame War's on Forums!" reporting in. Now it seems that the debate on whether computers may solve chess, yes dear listeners you heard those words right! They at the forum here are wondering if computers can solve chess, or in other words, take the mystery and fun out of chess! Yes folks that's their end goal! Any way back to the story, instead of debating with each other, many user have tried to collaborate in efforts to solve such question. They have so for presented with their efforts that it may be possible for such to happen, however it would be many life times from now. Also it seems that any debates on this subject have been sketchy. In terms of either side having much to argue about that is connected to this subject. We here at the station are wondering what you dear listeners think? We here at "Flame Wars on Forums!" are hoping all of our listeners have a great rest of the night! No on to our next story.....l

Come check us out! 

https://www.chess.com/club/flame-wars-on-forums-a-broad-casting-company

Their parents are paying...they just want to take the easiest/most fun courses they can that still allow them to graduate.  Then they find out that Game Theory is not "fun" ...

Wow, not a single personal attack or insult in this message from btickler! But it's completely vacuous! It's a reply to nothing that was said, and has zero meaning!!

...and now we have circled back to 4 pages ago...funny how all logic and discussion with you seems to be circular.

"Keep living in your dream world..."

just keep coming to my poker table. 

 Prove that state meant?

.. Semi-related o:

lol

you are welcome at my poker table without showing evidence of your advanced degrees. Your easy money is good enough for me. 

run well and play better

Thanks!

I don't understand why people are still arguing about this, in order to "solve" chess you would need a hard drive bigger than our planet, Why would they build that, and for anyone who says they could ignore the "bad moves", your not solving it if you ignore moves, and those moves might turn out to be good moves.

That's beside the point.