Live data from Hacker News

What Bayesianism taught me

lesswrong.com

1–10 of 14 posts

Re: What Bayesianism taught me

#2
Honestly this is where so much stuff falls flat on its face. People claiming they 'know' on either end of the nuts/pragmatic/scientifc spectrum.

'I think therefore I am' - cliche ofc. Only thing I can prove.

BUT. Prove a damn thing after that. If you can, you're not logically correct, you have no proof of anything outside of your existence. It looks real to you - that's your 'proof'.

We've all thought this at least once: It could be a simulation, you can't prove it's not.

You are choosing to believe you're living on earth, and there's a universe, and you're in it, and it's made out of stuff.

You are choosing this (perhaps sensible) scenario, but it is a choice. Because you may well be here reading this to yourself, from yourself, by yourself in the vacuum of your own imagination - and you wouldn't know.

Anyways... once you pass all that, you get to pick and choose what you prefer, and life becomes easier, and kinder, and more fun.. in my little experience anyway. Because you know everything you see and do is half-bullshit, and it's alllllright.

Re: What Bayesianism taught me

#4

Honestly this is where so much stuff falls flat on its face. People claiming they 'know' on either end of the nuts/pragmatic/scientifc spectrum. 'I think therefore I am' - cliche ofc. Only thing I can prove. BUT. Prove a damn thing after that. If you can, you're not logically correct, you have no proof of anything outside of your existence. It looks real to you - that's your 'proof'. We've all thought this at least o…

Whether I live in my imagination or in an external reality moderated by my senses, it would appear the same to me, and part of that appearing the same is that it seems to follow rules beyond my conscious will that I can guess at based on past experience.

I don't see how the fact that it might all be a dream makes the experience of pain any less painful or the experience of pleasure any easier to come by. I experience it all the same.

Re: What Bayesianism taught me

#5

Honestly this is where so much stuff falls flat on its face. People claiming they 'know' on either end of the nuts/pragmatic/scientifc spectrum. 'I think therefore I am' - cliche ofc. Only thing I can prove. BUT. Prove a damn thing after that. If you can, you're not logically correct, you have no proof of anything outside of your existence. It looks real to you - that's your 'proof'. We've all thought this at least o…

Whether I live in my imagination or in an external reality moderated by my senses, it would appear the same to me, and part of that appearing the same is that it seems to follow rules beyond my conscious will that I can guess at based on past experience. I don't see how the fact that it might all be a dream makes the experience of pain any less painful or the experience of pleasure any easier to come by. I experience…

It obviously does not help when you hit you little toe on the leg of a table, but it helps with more abstract problems. Realizing that philosophy can neither be right or wrong (only illogical) makes for a more relaxed life because you don’t have to worry about “the big questions”. You can just pick the answers you like.

Re: What Bayesianism taught me

#6

Is agnosticism actually a consequence of Bayesianism?

Depends on how you define agnosticism. To use a popular example, I am not agnostic about you having an invisible fire-breathing dragon in your garage despite not being able to put a flat 0% probability on it.

Bayesianism redefines the concept of knowledge, which under Bayesianism is represented as a probability. I have very strong knowledge that there is no god. If given sufficiently strong proof against that, I will update my beliefs about the issue -- but this does not mean the same as being unsure about it now.

Re: What Bayesianism taught me

#7
post #5

Earlier quoted context omitted.

Whether I live in my imagination or in an external reality moderated by my senses, it would appear the same to me, and part of that appearing the same is that it seems to follow rules beyond my conscious will that I can guess at based on past experience. I don't see how the fact that it might all be a dream makes the experience of pain any less painful or the experience of pleasure any easier to come by. I experience…

It obviously does not help when you hit you little toe on the leg of a table, but it helps with more abstract problems. Realizing that philosophy can neither be right or wrong (only illogical) makes for a more relaxed life because you don’t have to worry about “the big questions”. You can just pick the answers you like.

Do you have examples? Depending on what the Big Questions are I may or may not agree with that.

For example: It would be nice to believe in an omni-benevolent god that would take care of me after I die. But that's the sort of answer that I'd want to have confidence in - and I'd be aware that it's equally probable that there's an omni-malicious god waiting to torture me forever.

If I'd started off believing in evil-god then the knowledge that I had no information either way would be a comfort.

If I'd started off believing in good-god then it would be a burden.

Uncertainty in that respect neither comforts nor disturbs without reference to a baseline belief, it just is.

And when we start talking about things like morality, there is a sense in which there's a question as to what we're doing when we use moral language. However, most of the thought experiments concern real life examples anyway - outside of very specific areas of realist metaethics I don't think that's something that most people think about very much.

I'm... struggling to think of any big questions about reality that would affect my life. Ontology, epistemology - again most of the stuff I know is more about how people think about their experience - or can be converted directly into that talk with no loss of effect - than it is about esoteric metaphysics.

Re: What Bayesianism taught me

#8

Is agnosticism actually a consequence of Bayesianism?

http://lesswrong.com/lw/nf/the_parable_of_hemlock/

> The problem with syllogisms is that they're always valid. [...]

> The Bayesian definition of evidence favoring a hypothesis is evidence which we are more likely to see if the hypothesis is true than if it is false. Observing that a syllogism is logically valid can never be evidence favoring any empirical proposition, because the syllogism will be logically valid whether that proposition is true or false.

> Syllogisms are valid in all possible worlds, and therefore, observing their validity never tells us anything about which possible world we actually live in.

If a proposition is logically possible, there's a non-zero probability it's actually true. E.g. it's possible that God does not exist. It's also possible that God does exist and put dinosaur fossils in the ground to troll us. I say this without sarcasm.

Pure Logic(TM) isn't always pragmatic enough to suit our needs because it restricts its inputs and outputs to 100% certainties. In the real world, nothing is ever 100% certain. Maybe we have a lot of life experiences with gravity. Maybe we predict with seven 9's that objects will continue to fall down like they always have. But can we ever be truly-entirely-perfectly certain that tomorrow, gravity won't be around because God forgot to pay the utility bill? No, actually, we can't. We're just really confident because we've experienced evidence of gravity for our entire lives.

Likewise, you can never be truly certain that the diodes in your computer won't spontaneously reverse current. But we can still rely on diodes to work properly because we predict that such an event is super-ultra-extremely unlikely (has low probability) given the voltages humanity's electrical engineers decided to ram through our circuits.

Re: What Bayesianism taught me

#9

Is agnosticism actually a consequence of Bayesianism?

Depends on how you define agnosticism. To use a popular example, I am not agnostic about you having an invisible fire-breathing dragon in your garage despite not being able to put a flat 0% probability on it. Bayesianism redefines the concept of knowledge, which under Bayesianism is represented as a probability. I have very strong knowledge that there is no god. If given sufficiently strong proof against that, I will…

You shouldn't have any "knowledge" that there is no god.

The issue isn't that there is a small probability of god existing, but rather there is a vanishingly small probability of your particular brand of god being true versus all the other particular brands of god or gods being true.

The pudding in a Bayesian discussing is as so often in the normalizing factor...

Re: What Bayesianism taught me

#10

Earlier quoted context omitted.

Depends on how you define agnosticism. To use a popular example, I am not agnostic about you having an invisible fire-breathing dragon in your garage despite not being able to put a flat 0% probability on it. Bayesianism redefines the concept of knowledge, which under Bayesianism is represented as a probability. I have very strong knowledge that there is no god. If given sufficiently strong proof against that, I will…

You shouldn't have any "knowledge" that there is no god. The issue isn't that there is a small probability of god existing, but rather there is a vanishingly small probability of your particular brand of god being true versus all the other particular brands of god or gods being true. The pudding in a Bayesian discussing is as so often in the normalizing factor...

No. I have done quite a lot of observations on the world. Each and every one of those observations has produced a result congruent with a world in which no gods exist, and so has very slightly nudged my P(god) towards zero. Under a Bayesian philosophy, this probability is called knowledge.
Post reply on HN