Bucky

Karma: 2,797

Bucky 25 Jun 2018 14:30 UTC
3 points
on: Prisoners’ Dilemma with Costs to Modeling
Very interesting.
I think there may also be a Nash equilibrium with a mix of all 4 Bots, for a fairly small band of possible epsilon values.
d = 1 − 3e
f = ¹⁄₂
c = e
p = 2e − ¹⁄₂,
( ¼ < e < ¹⁄₃)
I haven’t proved this is a Nash equilibrium vs any modal agent, maybe you can.
This is unstable but I expect an evolution sim would obtain an oscillating pattern similar to the c,f,p equilibrium you find. In fact, when e = ¹⁄₃ this new equilibrium is the same as your original one.
Expected utility is 1+2e and so is less good than c,f,p for allowed epsilon values.

Bucky 21 Aug 2018 9:48 UTC
22 points
1
on: Historical mathematicians exhibit a birth order effect too
Would it be fair to say that any historical data on successful scientists/mathematicians will be over represented by firstborns due to primogeniture inheritance laws and customs? Historically those involved in the sciences mainly had to be independently wealthy and being a first born would tend to help with that with those born after more likely to have to work for a living. Maybe famous historical lawyers would tend to be under represented by firstborns?
I’d expect this to be a fairly large selection effect similar in size to the Less Wrong survey but presumably caused by a different mechanism.
Possibly a data set which would have more bearing on the question of birth order effects in modern times would be Fields medal, Abel prize, Turing award, Nobel prizes in Physics, Chemistry, Medicine and Economics in the last 30 years or so—I don’t have a great feel for how long ago the primogeniture inheritance thingy stopped being relevant but given an average Nobel laureate age of 59 this would mean people born since ~1930. These might be easier to find data on than Thales of Miletus too!!
What links here?
- Birth order effect found in Nobel Laureates in Physics by Bucky (4 Sep 2018 12:17 UTC; 61 points)

Bucky 22 Aug 2018 10:04 UTC
6 points
in reply to: Eli Tyre’s comment on: Historical mathematicians exhibit a birth order effect too
Thanks, that seems to rule inheritance laws out as a significant factor. I’m quite tempted to create that data set myself. Any objection if I use your analysis spreadsheet as a template?

Bucky 4 Sep 2018 15:53 UTC
3 points
on: Psychology Replication Quiz
Thanks for sharing this, great fun! ³⁶⁄₄₂ but I reckon if I’d done this a year or so ago (before reading the sequences/blue minimising robot etc.) I’d have got a lot less.
Incidentally, is it just me or was the “Don’t know” button a bit irrelevant. Even if I have no idea at all I can average 1 point by choosing an answer at random. If I have any idea at all my expected score for the question can only go up as long as I’m over 50% confident. If I’m less than 50% confident I should just choose the other answer!
I like the idea of having more than just yes or no but it would need to be more like Professor McGonnagal in hpmor—marking down for incorrect answers. Even a 1 point loss for getting one wrong would mean I had to be 67% confident to put in a guess. To allow players to enter your actual confidence level you could set it up like The £100k Drop.

Bucky 5 Sep 2018 10:13 UTC
1 point
in reply to: jessicata’s comment on: nostalgebraist—bayes: a kinda-sorta masterpost
If I had to run with the toolbox analogy, I would say that I think of Bayes not as a hammer but more of a 3D printer connected to a tool generator program.
Fasteners are designed to be used with a specific tool (nail & hammer, screw & screwdriver). Imagine, rather, you are trying to use an alien’s box of fasteners – none of them are designed to work with your tools. You might be able to find a tool which works ok, perhaps even pretty well. This is more like the case with a standard statistics toolbox.
If you enter the precise parameters of any fastener (even alien fasteners, whitworth bolts and torx screws) into your tool generator program then your 3D printer will produce the perfect tool for fastening it.
Depending on how well you need to do the job, you might use a similar tool and not bother 3D printing out a completely new tool every time. However, the fastening job will only be done as well as the tool you are using matches the perfect 3D printed tool. Understanding how your tool differs from the perfect tool helps you analyse how well fastened your work will be.
At times you may not know all of the parameters to put into the tool generator but estimating the parameters as best you can will still print (in expectation) a better tool than just picking the best one you can find in your toolbox.
This seems a little forced as an analogy but it’s the best I can come up with!

Bucky 5 Sep 2018 12:02 UTC
6 points
in reply to: Unnamed’s comment on: Birth order effect found in Nobel Laureates in Physics
Very good point. That would mean examining Nobel laureates in chemistry, medicine and economics might not give much extra information.
Literature laureates would, I presume, have significant intellectual curiosity so might give an indication as to whether the effect is specific to STEM subjects or more generalised curiosity/something else.
I’m not sure about Peace laureates – a negative result wouldn’t tell us much but a positive result would be interesting.
Any other suggestions for data which might be publicly available and point us towards one hypothesis or another?
We’re only going to get a good picture of what’s going on by someone deciding to do a proper research project on this but I feel like any additional data which people could collect would be helpful to point the project in the right direction.

Bucky 10 Sep 2018 15:17 UTC
1 point
on: On Robin Hanson’s Board Game
I wonder if something like this could be used in the workplace.
If we get an unexpected test result we all get together and come up with hypotheses as to what might be the cause. Each hypothesis has a column plus an “other” column and the game begins.
You’d need a robust method of adding new hypotheses (possibly which rewarded the hypothesiser—maybe giving them the first chance to purchasie stocks), plus a way of making it fair when new experimental evidence comes in (not just first come first served!).
I realise this is essentially just a prediction market at work but I think the visual elements would help people get it even if they aren’t familiar with the theory. I imagine it would help with considering what to test next and to make sure people keep an open mind.

Bucky 12 Sep 2018 8:36 UTC
5 points
in reply to: Richard_Ngo’s comment on: Unknown Knowns
Looking at the comments on the quiz link post, the average was 34.8/42. Using the same scoring system the prediction market would have got ³⁶⁄₄₂ (3 didn’t replicate which the market gave >50% credence to). If competent laypeople (some of whom by their own admission didn’t spend long on the task) can get 34.8, 36 doesn’t seem unreasonable.
I think the paper looks especially impressive due to the ability to give probability estimates but having done the quiz the ones I got wrong were amongst those which I felt least confident about.

Bucky 13 Sep 2018 7:51 UTC
1 point
in reply to: tenthkrige’s comment on: Birth order effect found in Nobel Laureates in Physics
So potentially ( Sports players ( Literature laureates / governors ( Physicists / Mathematicians ) ) ) ?

Bucky 19 Sep 2018 11:01 UTC
1 point
in reply to: jmh’s comment on: Birth order effect found in Nobel Laureates in Physics
One of the things I find most interesting is that the effect seems to be strongest for the rationality community.
I would suggest a sister theory to the intellectual curiosity angle that Scott mentioned. Eldest children spend their formative years without another child in the family to look up to. I would think that this would lead to, on average, a lower acceptance of information from authority.
This comes a bit my personal experience—I had an elder brother whose opinion I would take on as my own, even through my teens. I’d be interested to know if other younger siblings in the community have had a similar experience.
It would explain why the effect is so strong in the rationalist community. In Science and Maths it helps to challenge authoritative sources but this is actively encouraged, so the intrinsic effect needs to be less strong to get you to do it. The rationalist community is often challenging things which, culturally, aren’t supposed to be challenged which would give a higher intrinsic bar to entry.

Bucky 28 Sep 2018 11:28 UTC
19 points
on: Anti-social Punishment
It’s notable that, for countries where anti-social punishment is significant, the mean contribution across rounds doesn’t depend much on the level of anti-social punishment but more on the contributions in the first round; for the 7 countries with the most total anti-social punishment, their lines are all fairly flat.
Mean contribution and antisocial punishment (eyeballed from fig 2B in the report) aren’t correlated within the group of 7 (R2=0.08).
Mean contribution and initial contribution (eyeballed from fig 2A) are correlated within the group at R2=0.99!
So in countries with low rule of law you are stuck in whatever position you start in. Pity poor Istanbul which wasn’t really that bad at anti-social punishment but started at a low level and so were stuck there.
Imagine the experimenters had lied to each participant about what happened in round 1 to make it seem like everyone else was contributing more. Would the players stay at the high (made-up) rate of contributions for the rest of the 9 rounds?

Bucky 28 Sep 2018 14:07 UTC
1 point
on: give the evidence not the theory
I guess for obviously subjective things like tea this isn’t much of a problem.If something subjective feels objective it might be. Was this post inspired by any particular example?

Bucky 29 Sep 2018 19:32 UTC
1 point
in reply to: brown’s comment on: give the evidence not the theory
That is difficult when it feels obviously subjective to you but the other person takes it objectively. I tend to deliberately just say “I like tea” instead of “tea is great”. Your version sounds useful if you want to explore why you end up with different opinions.

Bucky 30 Sep 2018 19:09 UTC
1 point
on: The different kinds of confidence
I can relate to DaystarEld’s point of view. I wouldn’t frame it as changing my own confidence level but as the confidence level I display. In a job interview I will display more confidence in order to present my best side. Starting a new job I will display less so as not to come across as arrogant. But within myself I will feel equally confident in my own abilities regardless of circumstance. This feels kind of dishonest but until the world becomes more rational I can’t see an alternative which doesn’t impose significant costs on me.

Bucky 1 Oct 2018 9:01 UTC
1 point
in reply to: masasin’s comment on: The different kinds of confidence
I think the framing of 3 types of confidence is helpful but I think displayed confidence is something else entirely.
In the context of a job interview, the interviewee’s displayed confidence is often used (at least subconsciously) by the interviewer as a proxy for their competence. If I am up against a less competent individual who displays greater confidence (whether real or pretended) then in displaying my true level of confidence I am giving a signal to the interviewer that I am less competent than the other person.
Therefore I must pretend that I am more confident than I am, otherwise I lose out.

Bucky 1 Oct 2018 18:09 UTC
2 points
in reply to: cousin_it’s comment on: The tails coming apart as a strategy for success
That’s a great post,thanks for linking. I especially like the counterintuitive removing barriers to exit. I’m not sure that what I’m talking about works if applied to businesses—generally in business it pays to be tightly focussed. Being 10% better at what you do than your nearest competitor can mean that you get a 99% market share. The same may apply to employees in some very focussed professions but probably not many.

Bucky 2 Oct 2018 9:07 UTC
1 point
in reply to: masasin’s comment on: The different kinds of confidence
I think it’s mainly a matter of what I choose to focus on.
I won’t volunteer information on my confidence level and generally won’t be asked. In answering questions I will emphasize all the parts of the role I can do and what I’ve done in the past. I won’t volunteer the fact that I am not confident in a particular area as this would raise a huge unnecessary red flag for the interviewer. If I think I am below average at a certain skill I will say that I can do it (in a confident manner!), give what examples I can and leave it at that. This is true but doesn’t tell the interviewer the whole story.
A smart interviewer will dig down into the specifics and will ask more specific questions, meaning that you have to tackle your weak areas. In that case I will feel free to be more open as I will be more confident in the interviewer’s ability to accurately assess myself and the other candidates.
In my experience smart interviewers are rare.

Bucky 4 Oct 2018 15:32 UTC
3 points
on: Moloch in whom I sit alone
I’m torn between analysing this and just enjoying the prose. I think I’ll go for the latter.

Bucky 6 Oct 2018 16:27 UTC
7 points
on: The Valley of Bad Theory
A relevant experience: We spent a good few months tackling a motor problem where our calculations showed we should have been fine. In the end it turned out that someone had entered an incorrect input in one calculation and the test was far too harsh. All the time when we were having the problem the test engineers (read “people who have done a lot of iterative optimisation”) were saying it seemed like the test was too harsh. We wasted months trusting too much in the calculations. Sometimes the main uncertainty is that your model is wrong (or just inadequate) in which case iterative optimisation might win.

Bucky 10 Oct 2018 11:35 UTC
5 points
in reply to: Said Achmiz’s comment on: On insecurity as a friend
Thanks for linking those comments, I think I understand your thinking better now.
My understanding of Vladimir_M’s comments is that Scott’s proposed approach falls down if used indiscriminately rather than saying that it should never be used.
Here are 4 scenarios where I might modify my behaviour based on claims of harm (in descending order of likelihood of me modifying):
1= Someone I trust tells me that I have offended a 3rd party (without the 3rd party complaining about offence)
1= At least 2 people from different friendship groups (no cross-contamination) tell me I’ve offended them in a similar way
3. Multiple people from a single friendship group tell me I’ve offended them
4. Someone I trust tells me that I have offended them
Would you suggest that some/all of these should definitely not result in modification? Or is it that you think that generally people modify too easily and you want to encourage people to modify significantly less often?
I’m genuinely curious because if anything I suspect I do err on the side of being too accommodating.
P.S. I think the link to Scott’s concession goes to the wrong comment, it’s here if anyone is looking for it.