Posts
Comments
My first thought on reading this was that given that people tend to be overconfident in just about every other area of their lives, I would find it exceedingly surprising if it were in fact the case that people's estimates of their own attractiveness was systematically lower than the estimates of others. I notice that there isn't actually a citation for this claim anywhere in the article.
Indeed, having looked for some evidence, this was the first study I could find that attempted to investigate the claim directly: Mirror, mirror on the wall…: self-perception of facial beauty versus judgement by others.. To quote the abstract:
Our results show proof for a strikingly simple observation: that individuals perceive their own beauty to be greater than that expressed in the opinions of others (p < 0.001).
In other words, the phenomenon that you "explain" in this article is literally the opposite of the truth, at least for the people in that study.
Your strength as a rationalist is your ability to be more confused by fiction than by reality. Yes, surely some people under-estimate their own attractiveness, but if the explanation for this is cognitive biases which are present in everyone, how do we explain the people in this study who make exactly the opposite error? If you are equally good at explaining any outcome, you have zero knowledge, etc, etc.
Billy has the chance to study abroad in Australia for a year, and he's so mixed up about it, he can barely think straight.
Outside View - can anyone imagine a satisfying ending to this story that doesn't have Billy going to Australia?
meh, scratch that, I misremembered the quote as "most people disagree with you"...
I'd still bet that the majority of people who have a belief that meets all the criteria you suggest are probably wrong about that belief. For example, I think there's a reasonable case that most priests' religious beliefs would met your criteria, and it's clear that most priests are wrong (as long you you take priest to include holy men from all of the world's religions, it must be true).
I won't speak to the usefulness of the quote as a means for generating useful entrepreneurial ideas.
And conversely, things are allowed to just happen Because the Author Says So in fiction. When watching TV, I'll often ask "why didn't person X just do obvious thing Y which would have solved all of their problems for the rest of this episode?", to which my girlfriend's perfectly valid response is "plot reasons" (TV Tropes calls this the Anthropic Principle)
This seems to be the sort of cue that is much more reliable in fiction than in real life. In real life, not everything that happens has to be foreshadowed.
We're meeting up at 3pm on Sunday in Glasgow. It's not exactly planned as an hpmor wrap party (the organiser didn't even know hpmor was being updated again), but I expect well end up discussing hpmor, and any readers would be welcome.
It's not at all obvious to me that the failure mode of not looking for a better move when you've found a good one is more common than the failure mode of spending too long looking for a better move when you've found a good one - in general, I think the consensus is that people who are willing to satisfice actually end up happier with their final decisions than people who spend too long maximising, but I agree that this doesn't apply in all areas, and that there are likely times when this would be useful advice.
In the particular example I gave, if you've already found a move that wins a rook, then it's all-but irrelevant if you're missing a better move that wins a queen, as winning a rook is already equivalent to winning the game, but there are obviously degrees of this (it's obviously not irrelevant if you settle for winning a pawn and miss checkmate). This suggests you should be careful how you define a "satisficing" solution, but not necessarily that satisficing is a bad strategy (in the extreme, if your "good move" is a forced checkmate, then it's obviously a waste of time to look for a "better move", whatever that might mean).
It's also not always good advice. Sometimes you should just satisfice. Chess is often one of these times, as you have a clock. If you see something that wins a rook, and spend the rest of your time trying to win a queen, you're not going to win the game.
Serious question - why do you (either CFAR as an organisation or Anna in particular) think in-person workshops are more effective than, eg, writing a book, or making a mooc-style series of online lessons for teaching this stuff? Is it actually more about network building than the content of the workshops themselves? Do you not understand how to teach well enough to be able to do it in video format? Videos are inherently less profitable?
We will be organising another one relatively soon. If you pm me your email address, I'll include you in hype discussion
Just to be absolutely explicit, if you can't come, but would be interested in coming to a future meet up in Glasgow, please post here so we know you exist.
Did the survey. I don't know what cisgender means, but I assume that's me, as I'm definitely not transgender...
The Revelation Principle feels like one of those results that flip flops between trivially obvious and absurdly impossible... I'm currently in an "absurdly powerful" frame of mind.
I guess the principle is mostly useful for impossibility results? Given an arbitrary mechanism, will you usually be able to decompose it to find the associated incentive compatible mechanism?
I'm not sure I get this - If you're not allowed to compare apples to oranges, how do you decide which to eat? Is that the point this quote is trying to make?
If only one flower, we seek for nothing farther- what then if two or three, or more? Each successive one is multiple evidence- proof not added to proof,
Hard to tell out of context, but is this claiming that each successive flower is independent evidence? In general, it feels like the reasoner is missing some dependency relationships between bits of evidence here.
I was about 80% sure that 1159 was not prime, based on reading that sentence. It took me <1 minute to confirm this. I can totally be more than 99.99% sure of the primality of any given four-digit number.
In fact, those odds suggest that I'd expect to make one mistake with probability >0.5 if I were to go through a list of all the numbers below 10,000 and classify them as prime or not prime. I think this is ridiculous. I'm quite willing to take a bet at 100 to 1 odds that I can produce an exhaustive list of all the prime numbers below 1,000,000 (which contains no composite numbers), if anyone's willing to stump up at least $10 for the other side of the bet.
To be honest, no. There really isn't much more to it than is contained in the sixteen words above, or listening to one of Kaufman's TedX talks.
The First 20 Hours (Josh Kaufman):
Practice something for 20 hours, and you'll learn a lot. Don't worry about feeling stupid/clumsy.
How about this as a counter-example? This guy essentially got into Harvard because of one accident with a plagiarised essay when he was a kid (at least, that's the way he tells his story), and is now a member of faculty at Chicago. I think life outcomes might be more path-dependent than we like to admit.
http://www.thisamericanlife.org/radio-archives/episode/504/how-i-got-into-college
One thing that's unambiguous is that many ambitious high schoolers believe that where they go to college matters a great deal. My post is intended to address this audience.
It's possible that I misread, but I interpreted Swimmer963's point as saying exactly this - it really doesn't matter what you do in high school, as long as you get into the college you're aiming to get into. If this is what she meant, I probably agree - I don't think there is any one-semester high school course which can't be entirely learnt by a reasonably bright student in about 1 week of dedicated personal study.
Does anyone think they could win as the AI if the logs were going to be published? (assume anonymity for the AI player, but not for the gatekeeper)
The point being made by Gradgrind is much more basic: children should focus on Fact over Fancy.
ah, ok. I interpreted it as a preference for teaching Fact rather than Theory.
Re inbox zero: this paper seems to suggest it's a waste of time (and my experience concurs). How complicated is your folder structure?
Right - but did anyone not know that?
Am I missing something? Why is this quote so popular? Is there something more to it than "you can do harder sums with a pencil and paper than you can in your head"? Or, I guess "writing stuff down is sometimes useful".
I imagine this isn't your intention, but this does read a lot like "I think external review like AI experts would be good, but if we do that review, and don't liek the results, it's because we picked the wrong AI experts."
If you want smart kids like the rich folk, you should raise your kids like the rich folk raise their kids.
Is there any reason to believe this is true? I would guess Judith Harris would say no, and she's spent a lot more time thinking about this than I have.
The vast majority of people that show 90% or more correlation with me are concentrated in 2 areas of the world, New York city and California (SF Bay in particular), this is one of the indicators I choose for where I'll try to live.
Maybe you checked, but is it possible that the vast majority of OK Cupid users overall are in SF or NYC? This wouldn't surprise me at all.
Does Oz already know that he's a werewolf at this point? That would seem to bring "vampires exist" into the realm of plausible hypotheses.
Have you read the book?
My suspicion is that over 90% of it's worth is in an additional rule, which isn't one of these: "commit to practising something for 20 hours before starting to apply these principles". My guess - 20 hours of dedicated practice is just way longer than people tend to think it is, and you'd be surprised how much you learn in 20 hours without making an effort to do any of the rest of the 10 things.
The detrimental commute is driving
Is this actually true? I asked on this month's open thread, but didn't get a response. Has it been properly studied?
Will you able to be able to live near where you work in London? In Glasgow, you will almost certainly be able to afford to live in a nice place in the city centre where you can walk to work. People usually underestimate how much effect a long commute will have on their happiness (see e.g. here).
Also, you might consider trying this. If you're still unsure, chances are the expected value of the two options is pretty close, and you shouldn't worry too much about how you make the decision.
Disclaimer: I am in Glasgow, and would like to increase the population of LW-type people looking to make new friends who are here, so I'm probably biased.
this comment on the recent Reddit thread about intellectual jokes goes one better (and actually made me laugh out loud the first time I read it).
Apologies, I should have made this clearer (and will probably edit the original to do so). Commuting is terrible for the happiness of the commuter. The rest of the post should be interpreted in light of this.
As for the Freakonomics research - it seems quite implausible that the marginal commuter has a bigger impact by taking transit rather than a car (I seem to remember listening to an episode of Freakonomics radio about this discussion, and being disappointed by the lack of marginal analysis).
So, everyone agrees that commuting is terrible for the happiness of the commuter. One thing I've struggled to find much evidence about is how much the method of commute matters. If I get to commute to work in a chauffeur driven limo, is that better than driving myself? What if I live a 10 minute drive/45 minute walk from work, am I better off walking? How does public transport compare to driving?
I suspect the majority of these studies are done in US cities, so mostly cover people who drive to work (with maybe a minority who use transit). I've come across a couple of articles which suggest cycling > driving here and conflicting views on whether driving > public transit here but they're just individual studies - I was wondering if there's much more known about this, and figured that if there is, someone here probably knows it. If no one does, I might get round to a more thorough perusal of the literature myself now I've publicly announced that the subject interests me.
Is this even possible? How would someone know that a comment has been downvoted once it had been voted back up to 0 points?
Or more costly if you factor in the aggravation of dealing with the insurance company
Presumably if that was the case, she wouldn't have bought insurance...
So, since basically everyone in the world is overconfident, you can make them better calibrated just by making them come up with an interval and then doubling it.
What I've never really got is how you become accurately calibrated at the long tails. Are there really people who can consistently give both 90% and 95% confidence intervals? To me those both just feel like "really likely", and the higher the granularity, the harder it gets - note that a 98% confidence interval should probably be twice as wide as a 95% confidence interval. Are there people who have truly internalised this?
I personally like this two player calibration game, which I was introduced to by Paul Christiano at a meetup a couple of years ago:
- think of an unknown quantity (What year was the first woman elected to the US Congress?)
- Player 1 comes up with a 50% confidence interval (I guess, technically, this is a credible interval...).
- Player 2 chooses whether they want to take the "in" or the "out" side of the bet.
There's no need to choose a minimum width confidence interval (is there a technical term for that?) e.g. "before 1920" would be an acceptable confidence interval for the question given above.
The big advantage of 50% confidence intervals over 90% confidence intervals (other than that they make a nice easy structure for the game) is that you get much faster feedback. 20 trials can meaningfully tell you that your 50% confidence intervals are off in one direction or the other. 20 trials is enough to tell you if you're overconfident, but it can't tell you if you're underconfident.
The big disadvantage is that 50% confidence intervals somehow don't feel as useful as 90% confidence intervals. I'm not sure this is really true, as there's nothing special about 90% (by my reckoning 50% is about as far away from 90% as 90% is from 98%), but it feels true. Of course, it's pretty trivial to change the game so it works with intervals other than 50%, but you have to play longer, and it gets more complicated.
Not sure, as I'm not a native speaker of another language, but do most other languages use the same word for "should" in all three of those sentences? If not, it seems highly likely that it's just some sort of linguistic accident. If so, there might be something interesting worth pursuing.
The specific project I was evaluating had only gotten $800,000 out of the maximum $2m. Its strategy was to purchase the male students iPod Touches, the female students makeovers, manicures, and pedicures at a local beauty parlor, and all students were offered an additional iPod Touch or Makeover, respectively, if they passed the exam at the end of the current year.... only 25% (14/56) of the students targeted by the program had failed the reading exam in the first place.
$800,000/56 students = $14,000 per student. Those are some expensive iPod touches!
This is definitely true. General class of examples: almost any combinatorial problem ever. Concrete example: the Four Colour Theorem
Thanks, I was not trying the right combination of keywords.
(One could argue that these questions should probably be ignored and not investigated in depth - to paraphrase Teller, often magic is simply putting in more effort than any sane person would - but nevertheless, this is how things work for me.)
I can't find a source for this quote (and if it's from a longer interview, I think I'd probably like to read it), possibly because I'm not picking the right words to Google. Do you have a citation?
...that's about the last situation in which I'd expect people to rely on God
Does this cause you to doubt the veracity of the claim in the parent, or to update towards your model of what people rely on God for being wrong? I guess it should probably be both, to some extent. It's just not really clear from your post which you're doing.
I notice that http://www.miri.org is very definitely not a placeholder for a new Singularity Institute page. Have you managed to acquire it?
(miri.com seems as though it should be available, but not exactly entirely appropriate. Maybe better than nothing).
My experience with the GJP suggests that it's not. Some people there, for instance, are on record as assigning a 75% probability to the proposition "The number of registered Syrian conflict refugees reported by the UNHCR will exceed 250,000 at any point before 1 April 2013".
I am a registered participant in one of the Good Judgement Project teams. I have literally no idea what my estimates of the probabilities are for quite a few of the events for which I have 'current' predictions. Depending on what you mean by 'some people', you might just be picking up on the fact that some people just don't care as much about the accuracy of their predictions on GJP as you do.
I don't think it is an accurate reflection of the community. It certainly doesn't reflect my experience with the LW communities in Toronto and Waterloo.
It is also not an accurate depiction of the community in London or Edinburgh (UK). However, I think it is pretty close to exactly what I would expect a tabloid summary of the Berkeley community to look like, based on my personal experience. The communities in Berkeley and NY really are massively different in kind to those pretty much anywhere else in the world (again, from personal experience).
And, as Kevin says, it is remarkably nice - they could have used exactly the same content to write a much more damning piece.