Gogf
Indescribable
Perf's system explains how which box your take determines whether there is $1,000,000 in box B without violating causality or requiring time travel.
Newcomb's Problem
- Thread starter Erik Mesoy
- Start date
Perf's system explains how which box your take determines whether there is $1,000,000 in box B without violating causality or requiring time travel.
Then you are wrong, quite simply, if you appear to 1000 people and ask them what they will use to decide and most will use probability whether they are aware of it or not. Simply because all the other bs is unknowable and therefore irelevant. Thus given that situation Mises conclusions are perfectly reasonable and logical.
If your asking what would happen in reality, and what would happen if some people went off and discussed it for three millenia asserting all sorts of BS assumptions that may or may not be true, then yes in that circumstance probability is irrelevant, but that's not what the OP says at all. So I have no Idea who made you God? And why you got to decide whether you could consider probability in the issue, some form of determinism or randomness or not? Is this a discussion or are you the thought police?
Like I said - and you seem to be missing - it depends on the intial parameters (thus said parameters are the important consideration not the choice) the answer is always obvious if you set them up in such and such a way. I'm not sure why this is such an issue and why this thread has run so long, except some people obviously want to repeatedly state that given x: y as if it hasn't all been said before in various ways. Fine go for it, but don't tell anyone what they can and can't discuss thanks.
There is nothing wrong with what Mise said or anyone else who discussed probability, it is about probability given x, and not given y. You are not God and you do not get to decide what is applicable if you live in a reality that is based on such and such. Mine is based on probability and statistical manipulation of maths and science (it's what I study, it's what Mise and others studied) God knows what yours is based on? The Universe according to philosophical bs and vague unsubstantiated claims that have no basis in reality perhaps? Well welcome to the discussion, it's all arm waving bs. You don't get to pick what bs counts I'm afraid.
And btw if statistics become meanigless, then you are asserting that there is no relation to anything ever, which means that Omega's choice is not based on prediction and probability himself. Which takes the thread in a compeltely different direction. If Omega is somehow aware of information we are not, that doesn't change the fact that probability is involved, only that such information is not immediately apparent. Of course probability relies on not a single event but on many events to make a decision, this you have, this you will use if you have common sense. If not then you're obviously mentally ******** and are chosing based on blind choice, or assumptions you have no idea are real.
Place yourself in the position that some alien has just appeared and you know the OP. What are you going to use to decide? Think fast?
I could explain how you could not violate causality at all, in any circumstance you care to name in this experiment, but frankly stuff I've read is a bit highbrow for most. But say the word and I'll reel something out from the world of the cutting edge scientific hypothesis crew. But then I'm not making assumptions of how reality works, as if I know how it works now am I? And neither is anyone in science thank the lord.
Syterion
Voodoo Economist
I knew that. I have no idea how his quantum random number generator allows that at all.
Gogf
Indescribable
Gogf
Indescribable
Then you are wrong, quite simply, if you appear to 1000 people and ask them what they will use to decide and most will use probability whether they are aware of it or not. Simply because all the other bs is unknowable and therefore irelevant. Thus given that situation Mises conclusions are perfectly reasonable and logical.
If your asking what would happen in reality, and what would happen if some people went off and discussed it for three millenia asserting all sorts of BS assumptions that may or may not be true, then yes in that circumstance probability is irrelevant, but that's not what the OP says at all. So I have no Idea who made you God? And why you got to decide whether you could consider probability in the issue, some form of determinism or randomness or not? Is this a discussion or are you the thought police?
Like I said - and you seem to be missing - it depends on the intial parameters (thus said parameters are the important consideration not the choice) the answer is always obvious if you set them up in such and such a way. I'm not sure why this is such an issue and why this thread has run so long, except some people obviously want to repeatedly state that given x: y as if it hasn't all been said before in various ways. Fine go for it, but don't tell anyone what they can and can't discuss thanks.
There is nothing wrong with what Mise said or anyone else who discussed probability, it is about probability given x, and not given y. You are not God and you do not get to decide what is applicable if you live in a reality that is based on such and such. Mine is based on probability and statistical manipulation of maths and science (it's what I study, it's what Mise and others studied) God knows what yours is based on? The Universe according to philosophical bs and vague unsubstantiated claims that have no basis in reality perhaps? Well welcome to the discussion, it's all arm waving bs. You don't get to pick what bs counts I'm afraid.
And btw if statistics become meanigless, then you are asserting that there is no relation to anything ever, which means that Omega's choice is not based on prediction and probability himself. Which takes the thread in a compeltely different direction. If Omega is somehow aware of information we are not, that doesn't change the fact that probability is involved, only that such information is not immediately apparent. Of course probability relies on not a single event but on many events to make a decision, this you have, this you will use if you have common sense. If not then you're obviously mentally ******** and are chosing based on blind choice, or assumptions you have no idea are real.
Place yourself in the position that some alien has just appeared and you know the OP. What are you going to use to decide? Think fast?
I could explain how you could not violate causality at all, in any circumstance you care to name in this experiment, but frankly stuff I've read is a bit highbrow for most. But say the word and I'll reel something out from the world of the cutting edge scientific hypothesis crew. But then I'm not making assumptions of how reality works, as if I know how it works now am I? And neither is anyone in science thank the lord.
Sorry to say it Sidhe, but I really don't get most of what you're trying to say in that post. Here, though, is why I don't think probability matters at all:
1. If you cannot influence what is in the box, then it is absolutely, always better to take both boxes.
2. If you CAN influence what is in the box, then it is absolutely, always better to only take Box B.
Since there is no way to create a valid model for the probability that you can influence what is in the boxes, I fail to see how probability is relevant to this at all. Moreover, even if you COULD create that probabilistic model, basing your decision on the "expected value" is an awful idea. Choosing the box with the best "expected value" will get you the most money given a large number of trials. There is, however, only one chance here.
Obviously I was talking about the fact he has guessed 100/100 as was everyone else who was talking about probability. thus probability has everything to do with it. That's the friking turbot with lazer beams point of the whole question. Is there a probability involved in the unnering accuracy of the guesses or not. If you say yes there is a causal reason why alien is either a) accurate b) lucky or c) using prescience or whatever then you have probability, you can turn probabilistic outcomes off at a whim because you decided the universe isn't statistical. That's the point. That's why I said a dozen times now only the preconditions matter, the choice is always probabilistically determinable given the parameters. That is a lock down factomundi, unless you want to forgo causality, introduce hidden variables, or go whack out nuts. Which is your prerogative. But so far most people have been talking about things that physics in part agrees with, that don't break causality or x. So since physics oxygen is probability, who are you to deny it life? And how on Earth did you come to the conclusion that probability is irrelevant, in what situation is it irrelevant, think fast? Did some idiot tell you that that made sense, because if he did, he's a space maroon and should hang his head in shame.
If you have a probably out come, of 50.000000001/49.9999999999 then probability matters, nay it is the b all and end all of the discussion, but then the actual probabilities are determined by your preset preconditions. And can vary wildly according to that.
Gogf
Indescribable
You're misinterpreting my point. I'm not saying that probability doesn't exist, I'm saying that it doesn't matter in this case. There are two reasons for this:
1. You can't determine the probability that he is right with any reasonable degree of accuracy from him having been right 100/100 times.
2. Even if you could and could still come up with an "expected value," that knowledge would only be useful if you got to play this game hundreds of times. Even if the expected value of Box B was MUCH less than the expected value of taking both boxes, plenty of people would still be willing to take the risk.
Moreover, as I said before, either it's better to take both boxes or you can determine what is in Box B.
So what is the answer please?
For the third time:
suppose you know someone has been right 100 times out of a 100 (or even 1/1), what is the probability that he is right 50% of the time (or, equivalently, that he is guessing).
It turns out there is insufficient data to answer. Knowing a bit about Bayes' Theorem and prior probability makes it clear why.
I note that Mise also backed away from answering after claiming that it was the same thing as finding the probability that he is right hundred out of hundred given the fact that he is guessing. A conditional probability mistake, which aneeshm also made.
Mise
isle of lucy
I was probably wrong. No pun intended.
Anyway, as I (and Gogf and others) said, people who use probabilities to determine the truth or falsehood of Omega's predictive powers are missing the point entirely. That's true regardless of whether it's possible to do it.
That's right, because there is not enough information to use probabilities.
This isn't right. It isn't even wrong.
If probabilities were to be applicable, it would have to be to an entirely different problem. It is imaginable that, for such a different problem, probabilities WOULD yield the answer.
Mise
isle of lucy
Again, you're missing the point. If Omega's predicted 100/100 correctly, it would suggest that one of (a), (b) or (c) is true (read the thread). The only reason for someone to be a two-boxer is if they believe that none of (a), (b) or (c) is true. Hence, it's only rational to be a one-boxer.
So even though 100/100 tells us nothing firm about Omega's ability to predict our actions in the future, it is fairly strong evidence that (a), (b) or (c) is true.
What do you mean by again?
When did I miss the point previously?
If you read my posts, you'll note that I didn't even discuss the original problem except to note that people using probability to deduce Omega's theoretical success rate were making incorrect assumptions. You later agreed with this.
Again, this has nothing to do at all with anything I've ever posted in this thread. I havn't even argued for or against one or two-boxers, (a), (b) or (c), etc...
Again, my main claim was that there is missing data in the original problem to deduce what some people thought they were deducing.
My secondary claim (in my previous post) was that your statement that "people using probability are missing the point WHETHER OR NOT probability can be used" is meaningless.
To explain this again, if we were dealing with a problem where probability can be used, we would be dealing with an entirely different problem in the first place (since, as I explained many times already, it cannot be used in this way in the current problem) and saying that people are missing the point about a yet unknown problem is meaningless.
You may disagree with this and I'd like to know why if so but, as far as I can see, your last post was completely besides the point I was making.
Mise
isle of lucy
Your discussion about probabilities misses the point. In fact, most posts in this thread that mention the word "probability" miss the point.
It misses the point for two reasons, actually, but I'm only concerned with one of them.
So you're not actually talking about this problem? You're talking about a "yet unknown problem"?
Is it any wonder that I think you're missing the point?
Ya, because your point was irrelevant, and I was explaining to you what was relevant
Great Librarian
Warlord
The information that we have is that absolutely 100% of the time when Omega has chosen to put one million in box B he has been correct that the player will choose box B and that 100% of the time when Omega has chosen not to put one million in box B the player will choose both boxes. We also know that he puts the money there in the past, such that our decision about which box(es) to take does not affect the amount of money in the box at present. There is no conceivable way for Omega to cheat, as box A is always looked into or is irrelevant and box B is always looked into. Further, even though Omega has been observed to play the game, and accurately, with 100 people, we do not know the strength of the mechanism (nor the mechanism) by which he makes his predictions and we do not know how many of the 100 people chose both boxes or box B. There is no way we can make a meaningful assessment of Omega's prediction capabilities and the probability of his being correct except based on what we know.
Given the evidence we have it can only be rational to choose box B and only box B.
I don't understand why people are speculating about extra-explicit details as the game is closed. Making assumptions about the stuff does not produce an answer within the defined parameters and so any answer derived is meaningless as the game is defined.
Given the evidence we have it can only be rational to choose box B and only box B.
I don't understand why people are speculating about extra-explicit details as the game is closed. Making assumptions about the stuff does not produce an answer within the defined parameters and so any answer derived is meaningless as the game is defined.
Again, the only thing I did was correct a mistake a few people (including you) made in some calculations.
If someone based an argument on the fact that 2+2=5 then someone came in and simply corrected this particular mistake, would you claim that he missed the point?
Actually, you're the one who implicitly introduced a "yet unknown problem" with the following quote :
As I've explained a couple of times, since probability cannot be applied here, your last sentence must refer to a "yet unknown problem" where it IS possible to apply probability.
By your own criterion, it should be no wonder that you're "missing the point" (whatever that means).
You'll note that the only thing I said about this "yet unknown problem" was that it was meaningless (my earlier word) to talk about it. It seems you now agree yet it is what prompted your latest explanation of your "point" and my missing of it.
Actually, my only point was pointing out a mistake made by some (by you for example, again). If that was irrelevant, then it must be that the original point they were making was also irrelevant. That is of course completely possible. To be honest, I was only skimming through this thread and don't really care about the "points" being made (note that I never even referred to any of them). By training, I hate seeing things like probabilities misapplied and like to correct these things where possible.Ya, because your point was irrelevant, and I was explaining to you what was relevant
Mise
isle of lucy
I think you're confusing what I'm saying. It's irrelevent whether or not the probability analysis is applicable to the problem. Even if it is applicable to THIS PROBLEM, it makes no difference (at least not in the way that those using probability are suggesting).
Yes, that's exactly what I'm saying......... right here in fact:
Mise said:
What I am saying is that talking about probabilities in this way is irrelevant, and that's why I didn't respond to you. I don't really care whether the probability is calculable, because it is irrelevant. I thought it was clear enough...
That's fine. Since I don't care much about the original problem, it seems we have little left to discuss. I'll note that this wasn't always so since our "interaction" started with the following (incorrect) statement about probability:
which I corrected. Forgive me if the above comment gave me the impression that you DID care about this probability.
I'll throw a little more oil on the fire by stating that, again, the following
is vacuous, for the same reason as before. Not wrong, vacuous.
Since the probability analysis is NOT applicable to this problem, whatever claims you make out of the assumption that it is are trivial, vacuous truths.
You may as well claim that if the probability analysis was applicable to this problem, then 2+2=5. Logically that is also correct, yet content-less as your statement.
(See for example http://en.wikipedia.org/wiki/Vacuous_truth#Scope_of_the_concept)
Mise
isle of lucy
I cared about the analysis insofar as I thought it was mathematically correct. Beyond that, I don't care about it, since it is irrelevant to the problem.
It may be vacuous in a strictly technical sense, but it certainly isn't vacuous in the sense that most people commonly associate with the word "vacuous". It's actually crucial in understanding the problem, and it's this understanding that so many people have failed to grasp.
Somewhat ironically, whether or not the statement is technically vacuous is irrelevant, and makes no difference to my argument. I didn't argue against your assertion that the probability analysis was mathematically wrong, and for the same reason, I won't argue against your assertion that my statement is technically vacuous. Calling it a vacuous statement is a red herring.
I already said it depends how you set up the question. If there is no relation between Omegas ability to guess and causality and this leaves a 50/50 chance then yes in this absurd reality, probability is irrelevant specifically to that outcome, but it seems outside of philosophical debate causality is valid. And so are statistics. If it doesn't beg the question why he has guessed 100/100 then reality is put on hold to maintain a situation which is unlikely ever to exist except as a fluke. And in that case, yes there is a reason that probability shouldn't be used, except by the person making the guess, who has no idea whatsoever that Omega is such a jammy bastard and has no means nor can ever have any means to know what your choice is. Which to me seems fairly absurd, but knock yourself out. How you want to play around with causality is your look out, no matter how unrealistic or unlikely it is that anyone in the real world ever use such methods to chose. Seems to me to be philosophical semantics that wouldn't apply to a real world situation, but then I guess we've exhausted all the reasonable ideas, and are now grasping at the most unlikeliest of conclusions that only an idiot would make given x. After all the choser has no idea if there are any hidden variables that he doesn't know about, but would suffice to say assume there are, unless as said he was a moron. I think that was mooted quite early on, and since then the situations have been getting more and more unlikely/unrealistic.
Suffice to say though even if there is a 50/50 chance, saying probability is irelevant to the discussion is the most ironic statement I've seen made in a long while.
What did you use to come to that conclusion. And could you have come to a conclusion without using probability? Therefore how is it irrelevant, it underpins your question and its conclusions and requires a somewhat more than common or garden understanding of probability itself.I think the problem is people weren't precise in what they meant by probability being irrelevant so I jumped on that, not the idea that probability is irrelevant to the outcome or experiment given x, which is also false but for different reasons, and given a better expressed conclusion at least makes some sort of sense, if not really being particularly grounded in reality or without obvious caveats being a paradoxical statement.
Erik Mesoy
Core Tester / Intern
You are more complicated and harder to predict than a random number generator, so Omega can predict it too. If you go to get a quantum one, someone will have looted the boxes by the time you come back, and I think that that simply disqualify you like getting someone else to peek in the boxes would, anyhow.
Also: This is hereby an EPIC THREAD.
Similar threads
