[Merged] Immortality & Bayesian Statistics

[Giordano]

Dear Jabba,

Please let me know when you will do me and the others the politeness of responding to us. Thank you. Until then I will not go away as you might think I will. I have kids; I can put up with being apparently ignored for a long time.

 

[Slowvehicle]

Originally Posted by godless dave
Remember, in the scientific model, brains and observers are synonymous.

Originally Posted by Jabba
- Again, not quite.

Originally Posted by godless dave
How so?

Dave,
- But, you agree that identical brains would not produce the same observer. That's why brains and observers are not quite synonymous.

Good Afternoon, Mr. Savage!

Are you equivocating, or are you actually confused?

The fact that two brains, as "identical" as could be made, would each have a unique consciousness is a result of, an emergent property of, the fact that each brain is an individual brain.

Two different, "identical" Lego sets will produce two unique, but "identical" models.

I own two pair of the sadly-now-discontinued Chaco Tedinho Waterproof Hiking Boot. The are from the same manufacturing run, constructed of the same materials to the same specifications. In one sense, they are the "same", in that they are the same size, model, production run, style, and so forth.

However, I have no trouble telling hte two pair apart, when I wear them. Further, I venture to surmise that I could, blindfolded, determine whether any pair of the "same" boot was one of my pair.

Since what you vacillate between calling the "soul", the "observer", and the "self" (and which most of us call "consciousness") is a result of, an emergent property of, the brain; the "observer" ("self"/"soul"/"consciousness") is not, of course, synonymous with the brain. It is, however, unique to is one brain; and is not (nor has ever been demonstrated to be) shared by any other brain; nor has it ever been demonstrated to exist independently of the brain of which it is an emergent property.

If you have evidence to the contrary, now would be a lovely time to present it.

 

[Akhenaten]

Originally Posted by godless dave
Remember, in the scientific model, brains and observers are synonymous.

Originally Posted by Jabba
- Again, not quite.

Originally Posted by godless dave
How so?

Jabba, I think answering this question might shed some light on our misunderstanding:

Dave,
- But, you agree that identical brains would not produce the same observer. That's why brains and observers are not quite synonymous.

Hmmm . . .

Jabba is confusing "same" and "identical."

You probably realize that, but in a serious attempt to get him to do so, also, I offer the following:

Jabba, "Same" does not equal "identical" in the sense we have been using here.

Same = singular, one

Identical = indistinguishable but distinct

No worries.

I wonder how long it will be before Jabba again confuses the two and I have godless dave has an opportunity to quote your explanation.

6 hours and 50 minutes for the confusion to re-manifest and another 30 minutes for Dave's response, as it turns out.

We should have had a sweepstake on it.

 

[Giordano]

Hmmm . . .

​

6 hours and 50 minutes for the confusion to re-manifest and another 30 minutes for Dave's response, as it turns out.

We should have had a sweepstake on it.

I would have lost. I guessed it would be quicker than that, but Jabba may be having connectivity problems...

 

[Giordano]

Hmmm . . .

​

6 hours and 50 minutes for the confusion to re-manifest and another 30 minutes for Dave's response, as it turns out.

We should have had a sweepstake on it.

Is that an intrinsically finite, or an extrinsically finite lottery? Or an infinite one? The good news, if it will be an infinite lottery (given, of course, an infinity of time and resources) there will also be an infinity of winners. Even if only 1 in a million tickets wins. Oddly, this means I could still lose, given there will be a much "larger" infinity of losers than winners.

 

[xtifr]

The good news, if it will be an infinite lottery (given, of course, an infinity of time and resources) there will also be an infinity of winners. Even if only 1 in a million tickets wins. Oddly, this means I could still lose, given there will be a much "larger" infinity of losers than winners.

No, it's not a larger infinity of losers. Not only can you pair each winner up with a unique loser, leaving no extra losers unpaired (which means there are just as many losers as winners), but the naive way of doing so (the set of all pairs, [x, 10e6*x]) gives you winners paired with winners (e.g., when x=10e6), but no losers paired with winners, suggesting (incorrectly) that there are more winners than losers!

This is why infinity plays badly with statistics.

 

[jsfisher]

- I guess infinity is not a "number" -- but, as far as I can tell, it has to be an appropriate entry for some mathematical formulas...

Well, in the interest of full disclosure, infinity can be treated as a number. Not quite like a normal number, but basic arithmetic can be extended in a consistent way that includes infinity.

By the way, under that extension, 1/∞ is 0. Not just close to or really, really almost like, but exactly 0.

- Should we be considering the likelihood of my current existence given hypothesis A, or should we be considering the probability of someone/anyone's current existence given hypothesis A.
- My claim is that it should be the likelihood of my current existence given hypothesis A that we be considering -- and here, my current existence is the special case.

There's a problem with considering the likelihood of your existence. You see, you selected yourself because you already exist. Considering the likelihood of your existence is, well, already predetermined.

If you want to consider the likelihood of a hypothetical someone's existence, great. Knock yourself out, but you will never get near the result you seek.

If you want to consider the likelihood of the existence of any someone, and we have many examples to choose from, you are back to that same problem: You selected an existing something then ask about the likelihood it exists. Likelihood is a certainty.

 

[Loss Leader]

- I guess infinity is not a "number" -- but, as far as I can tell, it has to be an appropriate entry for some mathematical formulas...

Name 3.

Explain how they work and why infinity is the appropriate concept to use.

 

[Giordano]

No, it's not a larger infinity of losers. Not only can you pair each winner up with a unique loser, leaving no extra losers unpaired (which means there are just as many losers as winners), but the naive way of doing so (the set of all pairs, [x, 10e6*x]) gives you winners paired with winners (e.g., when x=10e6), but no losers paired with winners, suggesting (incorrectly) that there are more winners than losers!

This is why infinity plays badly with statistics.

No, this is why tongue-in-cheek works so poorly on the Internet. But you gave another reason I would not have thought of how to illustrate all infinities are the same. Thanks!

 

[jt512]

No, it's not a larger infinity of losers. Not only can you pair each winner up with a unique loser, leaving no extra losers unpaired (which means there are just as many losers as winners), but the naive way of doing so (the set of all pairs, [x, 10e6*x]) gives you winners paired with winners (e.g., when x=10e6), but no losers paired with winners, suggesting (incorrectly) that there are more winners than losers!

This is why infinity plays badly with statistics.

Infinity and statistics may have relationship issues, but they have agreed that even though there is a one-to-one corresponence between the set of winners and the set of losers, the probability of any member of the population being in the set of winners is still one in a million.

No, this is why tongue-in-cheek works so poorly on the Internet. But you gave another reason I would not have thought of how to illustrate all infinities are the same. Thanks!

Well, it is not true that "all infinities are the same." What is true is that all countably infiinite sets have the same cardinality (ie, are "the same size"). Uncountable sets have greater cardinality (ie, are "larger") than countable sets.

 

[jond]

Well, in the interest of full disclosure, infinity can be treated as a number. Not quite like a normal number, but basic arithmetic can be extended in a consistent way that includes infinity.

By the way, under that extension, 1/∞ is 0. Not just close to or really, really almost like, but exactly 0.



There's a problem with considering the likelihood of your existence. You see, you selected yourself because you already exist. Considering the likelihood of your existence is, well, already predetermined.

If you want to consider the likelihood of a hypothetical someone's existence, great. Knock yourself out, but you will never get near the result you seek.

If you want to consider the likelihood of the existence of any someone, and we have many examples to choose from, you are back to that same problem: You selected an existing something then ask about the likelihood it exists. Likelihood is a certainty.

Sorry to tell you that this post will be ignored by it's intended audience. The rest of us thank you for your efforts, though!

 

[Mojo]

Originally Posted by godless dave
Remember, in the scientific model, brains and observers are synonymous.

Originally Posted by Jabba
- Again, not quite.

Originally Posted by godless dave
How so?

Dave,
- But, you agree that identical brains would not produce the same observer. That's why brains and observers are not quite synonymous.

Jabba, if I have a pair of identical candlesticks, how many candlesticks do I have?

 

[Toontown]

There's a problem with considering the likelihood of your existence. You see, you selected yourself because you already exist. Considering the likelihood of your existence is, well, already predetermined.

That's quite a 'problem' you've discovered there. If you are correct, then all empirical tests of hypotheses are invalid. Because you see, the data used to empirically test a hypothesis already exists.

Your bad, bad news will be quite disappointing to all those people who have wasted so much time empirically testing hypotheses.

 

[jsfisher]

That's quite a 'problem' you've discovered there. If you are correct, then all empirical tests of hypotheses are invalid. Because you see, the data used to empirically test a hypothesis already exists.

Your bad, bad news will be quite disappointing to all those people who have wasted so much time empirically testing hypotheses.

That's all great, of course, but if you'd like to comment on my actual post rather than what you misinterpreted by post to be, that would be even better.

 

[Giordano]

Jabba,

Still no answer to the many criticisms here that have shown your theory to be wrong? Just curious...

 

[Toontown]

That's all great, of course, but if you'd like to comment on my actual post rather than what you misinterpreted by post to be, that would be even better.

OK, I'll try one more time.

There's a problem with considering the likelihood of your existence. You see, you selected yourself because you already exist. Considering the likelihood of your existence is, well, already predetermined.

"Does Jabba exist?"is not the question. Jabba's formula does not consider the likelihood that he exists. Jabba's formula considers the likelihood that he would exist IF hypothesis A is true.

If you want to consider the likelihood of a hypothetical someone's existence, great. Knock yourself out, but you will never get near the result you seek.

Can you explain why it's OK to consider how likely it is that something hypothetical exists, but not OK to ask how likely it is that an actual observation would be observed if hypothesis A is true, since hypothesis A gives the observation roughly a 1 in 400 trillion chance to be observed?

If you want to consider the likelihood of the existence of any someone, and we have many examples to choose from, you are back to that same problem: You selected an existing something then ask about the likelihood it exists. Likelihood is a certainty.

Again, Jabba's formula does not ask that question. There is no need to ask that question. The probability that some ranodm something exists converges on 1. But that doesn't mean the existence of some random something counts as evidene for or against some particular hypothesis.

Jabba's formula uses Jabba's self-observation as evidence (to Jabba) against hypothesis A, which says the expectation is roughly1 in 400 trillion that Jabba would be observing anything at all, ever, IF hypothesis A fully accounts for Jabba's self-observation. Except Jabba says it's 1 in infinity.

The formula cannot possibly mean anyhing to you unless you choose to use it on yourself.

 

[jsfisher]

"Does Jabba exist?"is not the question. Jabba's formula does not consider the likelihood that he exists. Jabba's formula considers the likelihood that he would exist IF hypothesis A is true.

P(Me) = P(Me|A)P(A) + P(Me|~A)P(~A)

so, yeah, it is part of Jabba's formula.

 

[Toontown]

P(Me) = P(Me|A)P(A) + P(Me|~A)P(~A)

so, yeah, it is part of Jabba's formula.

So, no. The left side of the equation is not the probability that Jabba actually exists (that's 1 irrespective of any numerical manipulations). The left side is Jabba's 'expected value'.

http://en.wikipedia.org/wiki/Expected_value

"the expected value (or expectation, mathematical expectation, EV, mean, or first moment) refers, intuitively, to the value of a random variable one would "expect" to find if one could repeat the random variable process an infinite number of times and take the average of the values obtained."

'Expected values' are typically used to test hypotheses, by mathematically comparing them to observed values. The 'expected value' is implied by the hypothesis. The observed value isn't implied by anything except your eyeballs.

 

[jsfisher]

So, no. The left side of the equation is not the probability that Jabba actually exists (that's 1 irrespective of any numerical manipulations). It's Jabba's 'expected value'.

We agree that the probability of Jabba's existence is 1. That would be P(Me) = 1. On the other hand, why you'd interpret an expression based on the Law of Total Probability as one for expectation, I have no idea.

By the way, what's the expected value for a constant? Let's say one that has a value of 1.

 

[Akhenaten]

OK, I'll try one more time.

Why?

"Does Jabba exist?"is not the question.

No, it's not. The question, if there even is one, is "Will Jabba exist forever?"

We seem to be getting further and further from that question with your every post.

Jabba's formula does not consider the likelihood that he exists.

And you don't find that intrinsically ridiculous?

Jabba's formula considers the likelihood that he would exist IF hypothesis A is true.

A formula which is conditional on the existence of the person attempting to use it is extrinsically ridiculous, at best.

Probably.