@allTheYud I'm curious, what is the FDT conclusion to this problem, from the FDT man himself?
English
truefire
394 posts
truefire
@truefire87
Game Designer and/or Programmer. Currently working on @RivalsOfAether.
Sumali Temmuz 2023
63 Sinusundan25 Mga Tagasunod
Utilitarianism supposedly yields an insane galaxy brained answer
Check details
Yep they're using causal decision theory instead of FDT
qualia receptacle@neocartesian
which button should absolutely impartial utilitarians press in the red-blue game? i ran the numbers: the answer may surprise you. a thread:
English
@IterIntellectus "The opinion that becomes more common as progress is made is wrong. Primitivity is evidence of correctness. I have depicted you as the midwit in the bell-curve meme, I win!"
English
pressing the blue button is the most luxurious of luxury beliefs
abstract altruism toward unspecified strangers is the last mental model to emerge in any human culture and the first to go when there is any form of struggle
it requires generations without any form of predator, institutional scaffolding none of which you built, and the kind of safety and material abundance 90% of the world lacks (and 100% of the world lacked since the beginning of mankind until like yesterday).
same kind of thinking as blank slatism or the kind that leads to universal mass migration, same operating system as "they just need better schools and diets"
the cost is always absorbed by people poorer than the believer. always.
mistaking naiveté and childishness for moral superiority.
you're not better, you're just a poor sweet child living in a fable and you should grow up
Max@minordissent
80% of the world’s population lives like this and you think they’re pressing blue?
English
There is a "woodchipper activation room". If more than 50% of people enter the woodchipper activation room, the rest of the world becomes a woodchipper that kills everyone who didn't go into the woodchipper activation room. Do you enter the wood chipper activation room?
The reason "two buttons" is a more meaningful framing is because it removes the inaction bias and reveals that "not doing anything" is a choice.
English
Right, because more than half of blue pushers are too stupid to connect button-pushing with choosing to die.
When you make them walk into a wood chipper or a blender or off a cliff, suddenly they realize “oh yeah, I could die if I choose this”
Ta-Nehisi Quotes@steady_drumbeat
Great example of where abstract game theory breaks down. Normal people will push blue, normal people won’t jump into a wood chipper!
English
I did a naive analysis of the 3-person case here that proves the first-order rule, rational agents should be able to extrapolate from this: x.com/truefire87/sta…
For a more in-depth statistical modeling analysis (not "perfect" game theory but a pretty good mathematical structure) take a look at the thread below. Worth noting that, while the math is sound, the model here assumes a uniform distribution of means for the potential beta distributions, which is highly unrealistic (and leads to a red conclusion). If you use a more reasonable distribution for beta means, such as a normal distribution, the conclusion tends towards blue. x.com/neocartesian/s…
truefire@truefire87
Others' Picks | Pick Red | Pick Blue Red + Red | 3 Live | 2 Live Red + Blue | 2 Live | 3 Live Blue + Red | 2 Live | 3 Live Blue + Blue | 3 Live | 3 Live -------------------------------------- Expected Value | 2.5 Live | 2.75 Live I see a lot of people saying "Game theoretically red comes out on top", and this is plainly false. If you are selfish, picking red is obviously optimal, as it provides 100% of your value (your own life) every time. You don't need game theory to figure that out. If you are fair (value every life equally, including your own) or altruistic (value every OTHER life) blue is optimal. I would assume anyone trying to make the game theoretical argument would consider themselves in the fair bucket. The actual right option is based on what the distribution of other people who pick red or blue. Given that the test is uncoordinated, we are assuming the distribution is roughly 50% (which is both the game-theoretically objective thing to do, and borne out in the data). The 3-person case above scales up to any number of people (with an increasingly smaller margin). You can run the math if you want.
English
English
Imo, most of Team Blue totally understands Red's arguments, but they disagree with the framing or assumptions.
otoh, most Red-pushers I've interacted cannot really articulate what the Blue position is. So they knock down a straw man.
The replies to this post will prove it.
English
@truefire87 setting high concentration gives you clustering around the mean for final propensities and a recommendation to press blue
English
which button should absolutely impartial utilitarians press in the red-blue game?
i ran the numbers: the answer may surprise you.
a thread:
English
@neocartesian This is a good analysis mathematically, but "given the uniform distribution for the mean" is doing a lot of work here. This is a pretty unreasonable assumption. Something like a normal distribution of means would be more believable.
English
in this setup, given the uniform distribution for the mean and a pareto distribution for concentration, red seems to win every time. however, the magnitude of the gain varies.
ultimately, both red and blue could be reasonable choices given your expectations of others' votes
English
@AudacityOfHoops @QiaochuYuan Yes, that's correct. There's a small margin around 50% (varies based on the population size) where blue is still optimal, and obviously anything over 50%.
English
@truefire87 @QiaochuYuan Those expected survivor number are only accurate if you assume other players pick blue and red at equal rates, though, right? Or am I misunderstanding the point?
English
i completely missed this discourse the first time so lemme try. i pressed blue without thinking about it, my gut reaction was "blue is prosocial and red is antisocial," on reflection this still seems right to me
galaxy-brain game theory arguments in favor of red are completely missing the point. as others have pointed out, empirically blue wins, and this is a test of theory of mind more than anything else. the actual outcome is determined by what everyone else who is not you actually does, not by what game theory says they should do
a certain kind of nerd thinks game theory is just the "correct" framework for reasoning about this type of situation and that is absolutely not true either and can be questioned on intellectual grounds and not just vibes. among other things game theory assumes every participant is perfectly selfish and perfectly """rational""" (and that this is common knowledge among the participants). this is just totally false as a description of the actual world! multiple parents pointed out that parents have to remember that this test includes their children. and obviously the vast majority of people have never even heard of game theory but they do know what selfishness and selflessness are
funnily enough there's LW stuff around exotic decision theories that's actually relevant here. one of them i would describe roughly as "when you make a decision you are choosing to live in a world where people like you make decisions like that" and i'd rather choose to live in a world where people like me are prosocial
"blue is prosocial and red is antisocial" is also a self-fulfilling prophecy, the more people who believe it the more true it becomes. so believing it is partly a bet on how much other people believe it, partly an act of hyperstition to make it true. few
Tim Urban@waitbutwhy
Everyone in the world has to take a private vote by pressing a red or blue button. If more than 50% of people press the blue button, everyone survives. If less than 50% of people press the blue button, only people who pressed the red button survive. Which button would you press?
English
@bitcloud @hyperliberalism "Communism" is when people care about other people, don't you know.
English
The "private vote" is doing all the heavy lifting here.
In reality there's no private vote. There are a series of decisions within a cultural framework, and those decisions don't occur instantly, but occur within a dynamic, reactive system.
If you return this back to a complex system you get high trust/low trust behaviours emerging exactly as they do in reality.
Low trust, red button societies collapse.
High trust, blue button societies thrive.
Tim Urban@waitbutwhy
Everyone in the world has to take a private vote by pressing a red or blue button. If more than 50% of people press the blue button, everyone survives. If less than 50% of people press the blue button, only people who pressed the red button survive. Which button would you press?
English
@ZPostFacto @eurocompel (*under the assumption that any individual life holds equal value)
English
@eurocompel A more reasonable and intelligent response than some of the "EV(red) > EV(blue)" 🧐 BSnkthers are saying
English
I like to think that I would press blue when the moment comes. I like to think that others would too. Maybe even some of the angry reds, when it comes down to it, would find greater humanity.
That said, who can say if I really have the heart. I've never been in a situation like that. I've only ever tested myself with lower stakes.
But this is an is/ought problem.
English
@truefire87 @choffstein @FlowTraderTM The difference is that I also see it as a moral failure, it is indeed shameful. And still.
The OP of the original poll had a very interesting follow up that illustrates this absolutely perfectly: x.com/waitbutwhy/sta…
Tim Urban@waitbutwhy
Suddenly standing alone in the room, I begin by imagining humanity banding together and blue winning in a landslide, and I feel a rush of pride. Red is the genocide button. Blue is the “save humanity from this nightmare” button. I know what kind of person I am. As my hand hovers over the blue button, I can’t help but imagine a gun pointing at my head with a bullet in one of the chambers. I feel a surge of fear shoot through my body. Then I think about all the other people staring at the blue button and thinking the same thing. Surely some of those who initially decided to press blue will succumb to the fear. It starts to feel like a gun with two loaded chambers. A stronger pulse of terror. The more I think about it, the more I worry about other people thinking about it. My heart races. Then I look at the red button—a gun with no bullets in it. A glorious feeling of relief washes over me. Will I hate myself forever if blue wins because enough others were better and braver than me? But don’t I owe it to my family to protect myself? One vote won’t change anything anyway, right? It’s all irrelevant because the mammal I live in has already made up its mind. I wince and press red.
English
Red looks selfish until you realize: if everyone presses red, everyone lives.
The "cooperative" answer requires coordination and trust.
The "selfish" answer just requires everyone making the same selfish choice.
Tim Urban@waitbutwhy
Everyone in the world has to take a private vote by pressing a red or blue button. If more than 50% of people press the blue button, everyone survives. If less than 50% of people press the blue button, only people who pressed the red button survive. Which button would you press?
English
@hitpeacebeuponu @0xCOSM0 You're right, every person is not going to simply pick red.
English
most people would try to justify picking red.
and you honestly cannot blame them, they aren't necessarily evil, just because they want to guarantee their survival.
this is just sad for people picking blue, cause they are thinking for the collective survival of everyone
betting with their life that complete strangers are good people
Tim Urban@waitbutwhy
Everyone in the world has to take a private vote by pressing a red or blue button. If more than 50% of people press the blue button, everyone survives. If less than 50% of people press the blue button, only people who pressed the red button survive. Which button would you press?
English
@hitpeacebeuponu @0xCOSM0 I need you to understand that if every person picked blue everyone would survive
English
@0xCOSM0 I need you to understand that if every person picked red everyone would survive
English
(Your post cut off at the end but I get the idea)
I find this to be an admissible perspective, compared to the average red advocate. But I do still consider it a moral failure. We should all strive to be blue pressers. A society can only thrive when the proportion of blue pressers is high enough. It's a bit of a commentary on the compounding destructive force of selfishness (as are most similar game theory problems).
English
@truefire87 @choffstein @FlowTraderTM Yes I understand I did use the exact same points before myself too. But I trust our flight or fight instinct and will to live more to overpower any sense of morality, in enough people, that it would push blue voters below 50%. And since I see the blue button as the suicide button
English
@AlexAegis @choffstein @FlowTraderTM *blue is correct both morally AND logically if you value human lives in general.
English
Red is "correct" in the strictly selfish sense, yes. But that's banally obvious.
If we value lives equally blue is strictly the correct option. (It's even the correct option if we place a higher but not categorically weight on our own life, up to a certain weight threshold depending on the sample size)
It's upsetting "obviously I should save myself over the rest of the world" is taken as as a given when people define what's "correct".
It is correct both morally AND logically if you value human lives in general.
English
@smolgob @BobVegana_ @choffstein @FlowTraderTM Ah but you've forgot the most important person (the only one who actually matters, by his own rationale) - him!
English
@BobVegana_ @choffstein @FlowTraderTM if red wins he kills one kid for sure if blue wins everyone lives
English
English
@AlexAegis @choffstein @FlowTraderTM I keep seeing red framed as "the correct answer that some people will just not get" but that's simply factually wrong. In any "pure logic" scenario blue is strictly better than red.
English