Re: Artificial intelligence in 1v1 diplomacy
Posted: Wed Mar 20, 2019 7:20 am
Someone just code this up already
https://www.webdiplomacy.net/contrib/phpBB3/
https://www.webdiplomacy.net/contrib/phpBB3/viewtopic.php?f=3&t=767
the CEO of DeepMind actually attended WDC 2017 Oxford, it's not impossible that this could happenbozotheclown wrote: ↑Wed Mar 20, 2019 7:46 pmIt would be interesting to see what AlphaZero could do with an imperfect information game like 1v1 Diplomacy. However, the estimate for the AlphaGo Zero hardware is $25 million, and AlphaZero likely costs more, so someone has to talk DeepMind into trying Diplomacy.
Did anyone explain 1v1 to them?captainmeme wrote: ↑Wed Mar 20, 2019 8:25 pmthe CEO of DeepMind actually attended WDC 2017 Oxford, it's not impossible that this could happenbozotheclown wrote: ↑Wed Mar 20, 2019 7:46 pmIt would be interesting to see what AlphaZero could do with an imperfect information game like 1v1 Diplomacy. However, the estimate for the AlphaGo Zero hardware is $25 million, and AlphaZero likely costs more, so someone has to talk DeepMind into trying Diplomacy.
Does anyone have some connections?captainmeme wrote: ↑Wed Mar 20, 2019 8:25 pmthe CEO of DeepMind actually attended WDC 2017 Oxford, it's not impossible that this could happenbozotheclown wrote: ↑Wed Mar 20, 2019 7:46 pmIt would be interesting to see what AlphaZero could do with an imperfect information game like 1v1 Diplomacy. However, the estimate for the AlphaGo Zero hardware is $25 million, and AlphaZero likely costs more, so someone has to talk DeepMind into trying Diplomacy.
I'm not sure there is a complete solution. Too much guesswork involved.
You can solve a probabilistic game. Nash equilibriums.Squigs44 wrote: ↑Fri Mar 22, 2019 4:54 amI'm not sure there is a complete solution. Too much guesswork involved.
The hard thing about diplomacy is its turn structure. In chess, or go, one decision is made at a time. White makes 1 move, then black makes one move, etc. In diplomacy, two players are simultaneously making up to 17 decisions that all interact with one another. So although there are fewer turns in diplomacy, each turn requires multiple complex interlocking decisions. Also, the fact that you are taking your turn at the same time as your opponent means you are working with incomplete information at every step.
You might be able to solve for the best move-set at any scenario with your given information using probability. But you can't solve for a "complete" solution in the sense that you can't guarantee that an AI will win every game it plays.Restitution wrote: ↑Fri Mar 22, 2019 5:17 amYou can solve a probabilistic game. Nash equilibriums.Squigs44 wrote: ↑Fri Mar 22, 2019 4:54 amI'm not sure there is a complete solution. Too much guesswork involved.
The hard thing about diplomacy is its turn structure. In chess, or go, one decision is made at a time. White makes 1 move, then black makes one move, etc. In diplomacy, two players are simultaneously making up to 17 decisions that all interact with one another. So although there are fewer turns in diplomacy, each turn requires multiple complex interlocking decisions. Also, the fact that you are taking your turn at the same time as your opponent means you are working with incomplete information at every step.
No, but that's not what solving a game means. Solving the game means that there exists an algorithm which can play perfectly.Squigs44 wrote: ↑Fri Mar 22, 2019 5:38 amYou might be able to solve for the best move-set at any scenario with your given information using probability. But you can't solve for a "complete" solution in the sense that you can't guarantee that an AI will win every game it plays.Restitution wrote: ↑Fri Mar 22, 2019 5:17 amYou can solve a probabilistic game. Nash equilibriums.Squigs44 wrote: ↑Fri Mar 22, 2019 4:54 am
I'm not sure there is a complete solution. Too much guesswork involved.
The hard thing about diplomacy is its turn structure. In chess, or go, one decision is made at a time. White makes 1 move, then black makes one move, etc. In diplomacy, two players are simultaneously making up to 17 decisions that all interact with one another. So although there are fewer turns in diplomacy, each turn requires multiple complex interlocking decisions. Also, the fact that you are taking your turn at the same time as your opponent means you are working with incomplete information at every step.
Nash equilibrium doesn't exist at every stage in this game.
Solving means to describe the optimal strategy. In this game, there is usually no "best move-set." Solving, in this case, means to describe the distribution of frequencies by which you should play each possible move-set. Then you randomize...
What can I say. In 50/50 guess for the win there is no Nash equilibrium. I think we don't use the same definition of "solve."Restitution wrote: ↑Fri Mar 22, 2019 4:41 pmIf diplomacy is solvable then there necessarily exists a Nash equilibrium for any decision
Even if AI plays against AI? Both would win?
The nash equilibrium is to do option A 50% of the time and option B 50% of the time, anything else would be exploitable. That is exactly what a nash equilibrium means... I do not mean to be rude but I think you might be thinking of a different concept, or have not been taught the concept correctly.
Pure sophistry
Sorry, but that is not what a Nash equilibrium is. A Nash equilibrium is a situation where both players made a choice, and neither player would want to change their mind even if they were told what the other player chose.Restitution wrote: ↑Fri Mar 22, 2019 6:29 pmThe nash equilibrium is to do option A 50% of the time and option B 50% of the time, anything else would be exploitable. That is exactly what a nash equilibrium means...