Practical Game Theory

[@nelltremagne2169]

Been watching y'all's lectures lately and I just wanted to add a small point of game theory that seems obvious, but a lot of people underthink it. 12:00 From the perspective of a designer, this is an important part of the second half of game theory- SELF INTEREST can't exist without INCENTIVE. When designing a game, if you're wanting to either utilize or avoid player utilization of game theory, you need to either add or subtract the element of gains from the equation. The reason most people (who aren't strategy gamers) don't find game like RISK or Chess enjoyable is because the only outcome is "victory", with a lot of minor risk-and-reward in between the beginning of the game and that final point. BUT those games are still unquestionably popular, almost ubiquitous. As y'all pointed out in your mechanics lecture, and I'm paraphrasing here, ["if you thought you were behind and then you win the game, you weren't really behind at all"]. Falsified incentive is what keeps you playing games like RISK and Chess. Those little victories that, in the grand scheme of the game, really don't mean anything are integral to keeping the flow of the game heading towards a victory on one side or the other. If chess, for example, was just a game composed of 15 Queens and one King, the pieces would have no intrinsic value to the player. If your opponent takes a piece, it doesn't mean anything to you, because you always have a body to replace it, until they take your King, and then the game is over. So victory (or endgame from design perspective) is the only incentive. You don't want your players to be looking forward to the END of your game, you want them to look forward to PLAYING the game. Therefore, the descending value of the pieces in a strategic sense become your incentive to say... gun for the Queen, or get rid of your opponent's Bishops, because that means a huge loss, strategically, to your opponent, which from a perspective of self-interest (and rationale, in a way), is always going to be satisfying to a player. Anyway, really enjoy your guys' lectures. Just wanted to add a random point for designers who might be scrolling the comments, lol.

[@eli0damon]

I am a math teacher. I have never seen game theory explained in this way. I think it is brilliant.

[@GeekNightsRym]

All our lectures are licensed Creative Commons 3.0. You're welcome to steal clips or show segments to anyone or mix them into lesson plans however you like!

[@N73B60]

They are the proof that the Ad Hominem fallacy is false.

[@nolanbarger]

Brilliant 👍 thank you

[@jonaskoelker]

I studied a bit of game theory at university. Scott gave an excellent definition of game theory in one of these talks, which I think every game theory professor should use, and which I'll inadvertently paraphrase a little due to poor memory: "game theory is the study of strategic interaction, which means how to make good decisions when the consequences of your decisions depend on the decisions of other people".

[@Larry21924]

This is nothing short of extraordinary. I recently stumbled upon similar material, and it was breathtaking. "Game Theory and the Pursuit of Algorithmic Fairness" by Jack Frostwell

[@chriscintron3561]

you guys give the best explanation of game theory. you make it easy and fun.

[@jkc2262]

😊

[@jkc2262]

😊😊

[@jkc2262]

😊

[@Danielle66666]

here in 2023 - thanks for the explanation :)

[@rameshsurulimuthu7397]

beautiful and Effective presentation

[@Thansk_]

-Figure out the other´s irrationality, watch for players who deviate from perfect play and try to exploit that. -pattern recognition is the key to winning games -As soon as you get good at a game, try to recognize other players who are clearly acting randomly and use that to your advantage

[@hellswhite383]

Thank you soooooooo much for putting this up! You guys are AWESOME!!!

[@uchenwogwugwu3567]

This lecture socialises game theory

[@jaimegarcia9088]

Guys, thank for these!

[@BobF510]

This content is a game-changer. A book with similar insights that I read prompted a profound shift in my life. "Game Theory and the Pursuit of Algorithmic Fairness" by Jack Frostwell

[@redsparowe2654]

Dude, I used to OWN that Superbike book when I was a kid. Nostalgia.

[@mjihiloussama9592]

Thanks for this brilliant and amazing introduction :) go ahead.

[@AlomgirkabirPolash]

Very helpful. thanks!

[@AndroidOO3]

These videos are amazing. Where are these guys from? How do they know this?

[@toddzelin4103]

All of this information is readily available in a number of books, in fact some of their examples are ripped directly from text that I've read.

[@GeekNightsRym]

Todd Zelin As is basically all human knowledge. I'd recommend "Characteristics of Games" by Richard Garfield and some of his colleagues as a good start. Focuses mostly on games, not "game theory," but is widely more practical. We draw many examples from it.

[@toddzelin4103]

Rym DeCoster Aye, I've read it. Been a fan of Mr. Garfield ever since I played MTG. :) Have to disagree with you though, a very large portion of human knowledge is only available in scientific literature.

[@RedRussianPedro]

Todd Zelin what are you disagreeing with?

[@shlamashona5123]

+Todd Zelin Nobody is saying they're geniuses, they're great at "teaching". Now stop trying to be an edgy bitch.

[@mattomattata2145]

Awesome video guys! Thank you :)

[@frage26]

lmao the lights at 28:25

[@ralevdotcom]

You rock! :) Thank you for sharing this useful game video. I'm playing it for a second time :)

[@Hikutachama]

I wish these were in parts

[@bboy32167]

I use a lot of these concepts on a regular basis cause i play magic the gathering a lot

[@aceylaurel996]

“We, the people” People have rights. Statutes do not say “the people shall”. Stand on your rights.

[@GeekNightsRym]

What?

[@ottodude555]

You think nobody noticed the Oglaf shirt you're wearing?

[@Cadellquimeric]

That's cool. Thank you!

[@confidentinterval3603]

I took a class called Game Theory last year. I can do the math :)

[@cookiesaregreat]

Great talk. Just not one thing "Humans have no ability to predict where something will land" This is actually incorrect. Scientists have determined that in many sports, such as ping pong, tennis, and yes, Baseball, it would be impossible to actually hit the ball UNLESS you predict where it's going to land mere milliseconds after the ball is struck or thrown.

[@midwesternbell]

THANK YOU! AWESOME

[@TimmacTR]

WHere can we find the slides to this?

[@bzsgzs]

They are like yin and yang - one with really long hair and the other's hair energy got sucked away.

[@Khrene]

I'm pretty sure they do the constrasting colors on purpose

[@Kuroyami0Fukaikuro]

What's the name of that 1812 war game?

[@EugeneBaldovino]

Amazing insight. Thanks guys!

[@Shaeress]

They should comment on the championships...

[@soapbxprod]

“There is not the slightest analogy between playing games and the conduct of business within a market society. The card player wins money by outsmarting his antagonist. The businessman makes money by supplying customers with goods they want to acquire ... He who interprets the conduct of business as trickery is on the wrong path.” -Ludwig Von Mises

[@NinjaFishleg]

The El Grande game map looks like Tamriel.

[@kaitengiri]

So early on, you say single-player games don't have multiple people and go off on the thing about the AI etc, etc, and that's all fine. But what about the idea that the other person in the game is actually the game developer? To expand on this, Mastermind, the game where you have to figure out the code the other player created within a certain amount of guesses based on limited feedback. Is that not a game? If it is a game, can we not consider that a single player game is perhaps an abstract form of Mastermind?

[@GeekNightsRym]

Per most readings of Game Theory, games that have a "bank" or other non-player participant can be considered multi-player if you treat that as a player with asymmetric payoff matrices.

[@kaitengiri]

Interesting. Thank you for the response.

[@blakebahner5237]

Those criticisms of game theory are common, but are false. Game theory only assumes people act rationally if you are assuming that the "perfect" result is the only thing you should do. Game theory studied properly involves all the possibilities and analyzing how those options influence the game.

[@user-vv7xo5jr5m]

Taylor Susan Thomas Donald Gonzalez Jessica

[@jonaskoelker]

For anyone who enjoys watching university people wearing their best Sunday ties for the occasion talk about game theory, kzbin.info/www/bejne/g2TSiIx-jtR1rM0 is a very fine 1h lecture analyzing the duel game-one of those goofspiel-like games that are more fun to analyze than play-with only very little math. For the math-savvy it might go a little slow, but that's what 2x chipmunk speed is for :-) The three key lessons are: 1. Dominance arguments: if it's better to do x than y no matter how your opponent then responds, you should do x without worrying about what your opponent will do. 2. Backward induction: if it's the last move of the game you can figure out what's best for you opponent to do; thus you can work out how to play in the second-to-last move. Your opponent can put themselves in your shoes and do the same calculation, so they can decide what to do in the third-to-last-move, etc. 3. How should your analysis change if you don't assume your opponent is playing optimally? It can't change dominance arguments but it can change backward induction arguments. The lecturer teaches an economics-oriented course on game theory, which you may or may not enjoy. It's at kzbin.info/www/bejne/pH6Wo4eLbpdqn9E.

[@LOCKEYJ]

I have no idea why he decided to throw food in his mouth whilst speaking but I feel threatened

[@ahappyimago]

Please try to get to the point sooner for future talks

[@DeerestAshe]

It's not just a theory its a GAME THEORY

[@wjeffcunningham]

A rock paper scissors AI that beats you 100% of the time. uh huh. Right. 🙄

[@williamhaffelder7032]

They might have good information but neither speaker lets the other finish a thought. It’s an hour of listening to them just interrupt each other.

[@brddks7629]

Hi nerds

[@Bldyiii]

Irony at its finest: theory on Rationality created by a schizophrenia patient.

[@joshuaginoza9446]

Four years later we know all of the possible combos to Chess and Go.

[@mehdimehdikhani5899]

this is not true.

[@chirilastefanchirila5990]

Can you stop speaking on top of other people. What do you think about this game

[@gotama570]

Let suppose 10 couples are having love? How many babies there will be? Just dont give me an answer nine and a half

[@leonardoraele]

I watch youtube to not have to read wikipedia and books and you tell me to go read wikipedia and books? Booooo!

[@octavioaraujo1674]

From a public speaking standpoint, the host on the right needs to stop cutting off his partner on the left. It is a little unprofessional from a public speaking standpoint.

[@theartofcompetition5965]

humans are not solvable, because of this.. games with high skill cap even at the highest level are not solvable because we are not robots there will always be edges to be had on a psychological level. specifically in a game where randomizers are not allowed. the more human element involved in a game the less "solvable" it becomes even more evident in a game such as poker where you can get reads on your opponent because the inability to control human mannerisms.

[@GeekNightsRym]

Try again. Chess has an insanely high skill cap, but no human has beaten a modern AI in a decade plus. Modern AIs win some variants of poker reliably now, and they're making rapid progress even on the holy grail of No Limit Texas Holdem. Poker is JUST as solvable as any other game, and there's no difference between a human's "psychological edge" and the same behavior in an AI. Within a few years, you'll see AIs that completely dominate humans in even no limit texas.

[@theartofcompetition5965]

I said humans in the very first word of the post. and you went straight to talking about vs a.i. (btw there is lebotus which has crushed some top poker players using a random balanced strategy) but I am talking about humans vs humans. you said yourself humans are very bad at acting randomly, I will add to that ..humans are not robots and have emotions that drive their actions no matter how hard we try to suppress them. (clearly this is where we disagree because you said "there's no difference between a human's "psychological edge" and the same behavior in an AI") this is just not true. a human has emotions that influence their decisions a.i. does not. and this doesnt even account for micro expressions. also... another thing to account for is stamina and things like this (which btw is the main reason lebotus beat poker pros) I guess if we really want to dig deep into this conversation im going to need you to define "solvable" if you simply mean mathematically solvable well then yes i can jump on board with that. because the human elements i mentioned all refer to the methods of executions, which are not solvable because they are always changing and evolving.

[@theartofcompetition5965]

why do you call texas holdem the holy grail lol i thought you didnt really play it

[@HDsharp]

A.I dont tilt thats how thay beat humans in poker..

[@theartofcompetition5965]

exactly. also ai doesnt get tired. the human brain gets tired very quick. even before the body. every try meditation? you are supposed to focus on your breathing... sounds simple enough...but your mind begins to wander after only seconds.

[@xranger707]

This Biff and Buffy BS goes on and on. WTF are they saying?