Crazy Thought Experiment/Logic Problem

[mitchiemasha]

The computer has never guessed wrong. That means it's predictions are very reliable.

If you pick A+B the computer would predict that and you'd get 1,100
If you pick B the computer would have predicted that as well and you'd get 1,000,000

If the computer has guessed correct every time up till now there's no reason it'd suddenly guess wrong now.
There's no "cheating" the computer. How would choosing both ever result in the computer not predicting exactly that?

Is the logic of taking both that you are at least guaranteed 1,000?

Yes that is 1 possible way to view the logic but our own definition of the logic doesn't get to decide how the universe actually works (or does it, but that's another story). Many people see it from the perspective you've wrote, others see it from the opposite. That's the point in the question, it shows which category you fall into. Personally i see both sides and accept that both may be true. It's impossible for todays current science to prove either side. I might be tempted to have a favourite, what i'd like to be true but i wouldn't swear by either. I'm like that with many things.

[Gold Bearer]

The computer has never guessed wrong. That means it's predictions are very reliable.

If you pick A+B the computer would predict that and you'd get 1,100
If you pick B the computer would have predicted that as well and you'd get 1,000,000

If the computer has guessed correct every time up till now there's no reason it'd suddenly guess wrong now.
There's no "cheating" the computer. How would choosing both ever result in the computer not predicting exactly that?

Is the logic of taking both that you are at least guaranteed 1,000?

The logic of picking both is simple. If there's 1,000,000 in box B then you'll get 1,001,000 if you take both and 1,000,000 if you only take box B. If there's 100 in box B then you'll get 1,100 if you take both and 100 if you only take box B. Either way, you'd be better off taking both. The most common argument is that the above is the only rational way of looking at it and this setup simply rewards irrationality. It seems paradoxical though to claim that all the people that went home with less money are the rational ones.

It's impossible for todays current science to prove either side. I might be tempted to have a favourite, what i'd like to be true but i wouldn't swear by either. I'm like that with many things.

"I try never to understand anything, it's called an open mind."- The Doctor. You could argue that it will never really be possible to answer. The question is that good. It's easily the best one I've heard and it's actually really simple, just complicated to fathom.

[Goblin-King]

But in your explanation you simply refer to box B having either 100 or 1,000,000.
While that is true, you don't seem to address the fact that the content is based on a reliable prediction.
The relationship between the content and the prediction is extremely important.
If the content was randomly put there I'd agree; take both boxes. But it's not random. And you even know what's in the box because it's based on your choice.

And again, there's no cheating the computer. If you choose only A. Then that's what it predicted.
If you choose only A, but then in the last second pick A+B then THAT'S what it predicted.
So it kinda boils down to a question of determinism. Can you diverge from the prediction?

This comic seems to be in place here:


Wiki DID say most had a hard time seeing the other view! I'm trying though...

[Count Mohawk]

Hey guys, I just thought of an interesting wrinkle to add to Gold Bearer's first logic puzzle.

What if you are presented with the two boxes as normal, and told to take either Just Box B or Boxes A & B, and are told about the computer that has never been wrong about what it predicts, but are also told the super computer has only ever been used once? Would that change your answer?

(Now note that there's nothing in Gold Bearer's original scenario that precludes the possibility of this being the actual case, save that you are not told how many times the super computer has been used...)

[Goblin-King]

It has NEVER been wrong in all of the 1 time it has been used!
My answer would be to tell them to test their product properly!

But seriously, in this scenario the computer goes from being reliable to being unreliable.
If we can't trust the computer to predict accurately then it's just a coin toss and A+B is preferable.

[mitchiemasha]

You could argue that it will never really be possible to answer.

I've said the same on other things.

[mitchiemasha]

It has NEVER been wrong in all of the 1 time it has been used!
My answer would be to tell them to test their product properly!

But seriously, in this scenario the computer goes from being reliable to being unreliable.
If we can't trust the computer to predict accurately then it's just a coin toss and A+B is preferable.

Which is the point. It doesn't matter what the computer predicts... In the other model of the universe, it couldn't possibly know for sure what you would pick, the Chaos Theory prevents this... Remember, what ever is in the box, will always of been in the box and will always be in the box regardless to your personal choice, on the day that you make it (edit: depending on which model the universe actually is). It's Determinism v Freewill, if determinism wins, a computer could model the universe and never be wrong, it may take all the energy in the universe to make it 100% accurate, the butterfly effect and all that... but if freewill, the Chaos theory is correct, there is always a margin of error.

In many science fiction film we have computers that predict the future... They predict a crash in stocks so every one sells the stocks, causing the crash, they predict plague, so every one is rounded together, causing the plague, they predict war so war is started. In the freewill model, just because one predicts what actually happens doesn't mean they knew it would happen, they simply guessed right. A machine making predictions, therefore can be a very dangerous thing, remove the machine and those things don't happen, it could also predict the opposite, creating good things.

The obvious choice by many would be to pick B and hope the computer was right, never proving it wrong so the test results would actually be flawed. It would take the first brave soul to try take both boxes to see, however, even a simple algorithm would be able to tell you were the type to do that after asking a few questions.

The point of the Chaos theory is... We may know exactly how sand will blow in the desert but we can't predict where each grain will be. A Determinist could then argue, because your prediction machine doesn't have all the variables, only the universe itself has all the variable. Followed by the non Determinist to say, yes but that machine itself would be come a new variable and as all energy simply transforms, the maths prevent it from ever existing, it would need to factor in it's own existence, then factor in its own factoring in of it's existence, and so on, creating an infinite loop on needed energy. The collapse of the universe... BOOM! big bang all over again, lol!!!

[Gold Bearer]

This comic seems to be in place here:

Wiki DID say most had a hard time seeing the other view! I'm trying though...

I’ll just argue from that point of view.

Whatever box B has in it has already been decided and you’re given a choice between taking just that box or that box plus 1,000. I’ll have whatever’s in box B plus the 1,000 please. That fact that something has made a prediction is completely irrelevant. That prediction has already been made and whatever it was, you’re now better off taking both boxes.

The point of the Chaos theory is... We may know exactly how sand will blow in the desert but we can't predict where each grain will be. A Determinist could then argue, because your prediction machine doesn't have all the variables, only the universe itself has all the variable. Followed by the non Determinist to say, yes but that machine itself would be come a new variable and as all energy simply transforms, the maths prevent it from ever existing, it would need to factor in it's own existence, then factor in its own factoring in of it's existence, and so on, creating an infinite loop on needed energy.

The simulation hypothesis, or my version of it at least. They only thing capable of truly simulating every variable in the universe would be something that was indistinguishable from the universe itself. So are we in a simulation, and is a universe that isn't a simulation even possible? What is a simulation other that a bunch of equations? Maths isn't created, it's discovered, so those equations don't need to be formulated to exist, they're already there, so here we are.

[Goblin-King]

I agree that the content of the box doesn't magically switch around on the fly depending on what you are considering this very instant.
The computer has made it's prediction and placed the content of the box before you enter the stage. That part is static.

But isn't the point that this computer would indeed be able to predict the butterfly's hurricane on the other side of the world?
Denouncing the existence of such a computer is kinda dodging the question in my opinion. Instead of a computer it could also be God.
*Something* is able to make an accurate reliable prediction.

Also, obviously you could predict everything if you had all the variables and processing power.
Even if you accept this as fact, it's useless as those two criteria can never be met.
And while I personally believe it to be true, the near infinite amount of variables makes life approximate randomness and free will to a degree we certainly can be content with.

[Wolf]

Just box A. Due to logic it's a coin flip if box B will have 100 or the higher number but I know for a fact Box A has 1,000 credits so I would just take that.