Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Bayes's Theorem: What's the Big Deal?

blogs.scientificamerican.com

Post reply on HN
Print view View original

251–260 of 267 posts

Re: Bayes's Theorem: What's the Big Deal?

#251

Post by marcosdumay »

hyperpape wrote: post #161
marcosdumay wrote: post #149

Earlier quoted context omitted.

Maybe that the natural frequency "normatized" for some people may be completely different from the actual frequency that apply to them, because the people that researched the frequency did their grouping badly.

I'm not sure if that's relevant. It seems like either you know the actual frequency for their subgroup, in which case you redo the numbers in the example, or you don't in which case you can't accommodate that regardless of how you calculate. A natural frequency is different way of expressing and reasoning about the same information as percentages.

Well, me neither. If you get the frequency wrong, you'll also get your priors wrong. Natural frequency does not compose well, but that is another problem.

That is just what I understood from the GP.

Re: Bayes's Theorem: What's the Big Deal?

#252

Post by XFrequentist »

kybernetikos wrote: post #250
XFrequentist wrote: post #248

Earlier quoted context omitted.

No - the precondition is that agents share common knowledge of each others' priors ( not that they have the same priors).

Here's the paper I read: http://www.ma.huji.ac.il/~raumann/pdf/Agreeing%20to%20Disagr... Here's how it starts: "Theorem: if two people have the same priors..."

Whoops, seems you are correct. I'd misunderstood this, my apologies.

Re: Bayes's Theorem: What's the Big Deal?

#253

Post by woodchuck64 »

jameshart wrote: post #194
philh wrote: post #190

Earlier quoted context omitted.

You're not engaging with the claim as it was meant. In context, no human being has ever been "much more intelligent" than me. Not in the same way that I am "much more intelligent" than the monkey von Neumann. You might decide that this means edanm goes off the rails at step four, instead. But you should at least understand where you disagree.

I'm still not sure you could assume ultimate power and achieve everything you desired if you were the only hacker news reader on a planet of 8 billion monkeys.

> I'm still not sure you could assume ultimate power and achieve everything you desired if you were the only hacker news reader on a planet of 8 billion monkeys.

I would think it relatively easy for a human armed with today's knowledge and a reasonable yet limited resource infrastructure (for comparison to the situation of an unguarded AI) to quite easily engineer the demise of primate competitors in the neighborhood. Set some strategic fires, burn down jungles would be the first step. "Fire" might be a metaphor for some technology that an AI might master that humans don't quite have the hang of yet that can be used against them. For example, a significant portion of Americans seem way too easily manipulated by religion and fear, an AI-generated televangelist or Donald Trump figure might be a frightening thought.

Re: Bayes's Theorem: What's the Big Deal?

#255

Post by cbd1984 »

RodericDay wrote: post #30

I think Stephen Bond did some excellent takedowns of the identity politics that has arisen around Bayes' Theorem back in the day. I wonder where he's at these days. The Cult of Bayes' Theorem http://laurencetennant.com/bonds/cultofbayes.html > One of Yudkowsky's constant refrains, appropriating language from Frank Herbert's Dune, is "Politics is the Mind-killer". Under this rallying cry, Lesswrong insiders attempt to…

This is why people don't take philosophers seriously: They apparently can only attack strawmen, and badly at that, using emotion in place of reasoned arguments.

Re: Bayes's Theorem: What's the Big Deal?

#256

Post by cbd1984 »

cwyers wrote: post #5

> In many cases, estimating the prior is just guesswork, allowing subjective factors to creep into your calculations. You might be guessing the probability of something that--unlike cancer—does not even exist, such as strings, multiverses, inflation or God. You might then cite dubious evidence to support your dubious belief. In this way, Bayes’ theorem can promote pseudoscience and superstition as well as reason. Oh…

> my "ASK ME ABOUT ROKO'S BASILISK" novelty t-shirt idea.

Why should we take you seriously when you resort to such cheap attacks?

Re: Bayes's Theorem: What's the Big Deal?

#258

Post by cbd1984 »

jsprogrammer wrote: post #257
cbd1984 wrote: post #254

Earlier quoted context omitted.

Believe it or not, the Heisenberg Uncertainty Principle has nothing to do with measurement.

Is someone claiming otherwise?

Lots and lots of people, especially people who don't know about conjugate pairs or the Fourier transform.

Re: Bayes's Theorem: What's the Big Deal?

#259

Post by manish_gill »

loup-vaillant wrote: post #226
manish_gill wrote: post #220

Earlier quoted context omitted.

If your ideas can stand on their own merit, you can defend them against criticism without resorting down to saying things like "oh it's a pseudo criticism and cherry picking facts". You're the one calling it bullshit. You even admit you wouldn't want to refute specific accusations. And then go on to call it "malicious bullshit". So, basically what you're saying is that we shouldn't listen to this criticism (which pro…

Let's try: as far as I know, this bit: Under this rallying cry, Lesswrong insiders attempt to purge discussions of any political opinions they disagree with . Having hanged around LessWrong for quite some time, I'm pretty sure this point is false. How am I to show it however? Dump the whole site to show the absence of such purges? And how the inevitable counter-example? Quite a lot has been said there, we're bound to…

Alright, lets see

> Having hanged around LessWrong for quite some time, I'm pretty sure this point is false. How am I to show it however? Dump the whole site to show the absence of such purges?

To refute the above statement, only one statement to the contrary would be a good start - it would provide me with enough evidence to consider your side of the story and think that perhaps the critique isn't completely fair.

> Quite a lot has been said there, we're bound to find unacceptable behaviour somewhere.

This can be said for any popular community. Does that mean we should not critique them? Or that some how invalidate any criticism against them? No.

The thing you're missing here is not the number of counter-examples, but a general cultural trend. HN, for example, is a tech oriented community, but political discussion isn't anathema here. I can show this by reading a few posts and noticing how people are comfortable discussing differentiating political opinions here. It's not perfect argument, but it's a good start, and goes a long way towards making me appreciate the environment.

Here, there is an important role played by so called "leaders" in community and how tolerant they are as well, by the way. And I know for a fact that Elizer isn't one of them. Then again, what else do you expect from a guy who tries to edit his own Wikipedia page.

> If the criticism can indeed apply to all groups, then it effectively applies to none, because it doesn't provide a way to discriminate different groups.

I'd happily accept this if you specifically mention which criticism in the parent link can be applied to all. The critique is very, very specific. It starts by defining what Bayes Theorem is, what LW and their ilk think about it, more of their pseudo-intellectual talk, the cult of personality, the AI apocalypse doomsday scenarios, etc etc.

Re: Bayes's Theorem: What's the Big Deal?

#260

Post by manish_gill »

sawwit wrote: post #242
manish_gill wrote: post #235

Earlier quoted context omitted.

> I cannot judge to which degree these theses are bullshit. Go the link, go to references, read about them. I'll outline the gist: Most of what Yudkowsky says is extremely sci-fi, no real basis in scientific fact, but stretching the current technological progress to the point where his opinions on things (stuff like transhumanism, singularity) can be justified. What he's preaching isn't science. Certainly not rigorou…

I will look into Farnham Street's Blog, thanks. > What he's preaching isn't science. I don't buy into the necessity that everything has to be peer-reviewed in the old fashioned way. There is peer-review happening in the comments to some extent. I don't take a fancy to dismissing any radical ideas as pseudoscience. It's just the outer fringe of hypotheses that need to be tested against reality, and as long they are ap…

> There is peer-review happening in the comments to some extent.

Lol. I guess we don't need college education as well then, there's education happening in the comments to some extent. We don't need traditional means of news, there's news happening on Twitter to some extent. I could go on with analogous line of reasoning.

Don't get me wrong, I'm not 100% in favour of the traditional education model as well, but peer reviews exist for a reason. You and I are not experts in these fields. We rely on the expertise of people who have made it their business and life to study these fields based on a rigorous method. Would you try out homeopathy had it not been rejected completely by doctors and scientists but someone on a forum told you it worked for them? What if someone wrote a very long article with fancy words (like LW tends to do) explaining how and why it works (they exist, I assure you)? Would you try it then?

> I don't take a fancy to dismissing any radical ideas as pseudoscience.

Sure, I'm not saying we should be against radical ideas. That's how scientific progress happens. I'm against LW ideas, for which there is no basis in reality as far as we know based on our current understanding of science.

> I differentiate within these communities between what is valuable and what is not.

Indeed. But I'd rather the community's entire existence not depend on bullshit.

> At the risk of seeming indoctrinated to you, ...

a) Keywords: "If", "Seems" b) Tons of assumptions in that scenario you laid out. If you can't see it, I'm sorry but you're already too far gone. c) Watch some MIT lectures on computer architectures about how the trend of Moore's law has already radically shifted and is flatlining.

Basically, what you've done is precisely the kind of utter crap that LW perpetuates. "If x keeps happening" without providing any reason as to why that would be true. Make some ridiculous simplifications "complexity is limited by ___", nature often does __ because ___. You basically don't provide any rational reason for why you think AI will be super intelligent and even if it were, why that would be risky. You pick numbers out of a hat (10-40 years to come).

Yes, you look pretty well indoctrinated from where I'm sitting. But I hope you see the many (so many) flaws in that last paragraph of yours (it honestly made me laugh out loud :p)

Predicting the future is hard business -- be it the stock market predicting what happens tomorrow, or weather forecast for the next month. It's presumptuous and hella stupid if you think you can predict where Science and Technology will be x years from now.

TL;DR: Stahp.

Post reply on HN