Saturday, September 05, 2009

Some favorite posts

NEW!  Podcast show Manifold  -- also on YouTube.

I started writing this blog in 2004 (it has had millions of visitors!), and by now the content is a bit unwieldy to navigate, even with labels and search. I thought I'd make a list of some of my favorite posts and topics. Please suggest other posts to add to the list!

Pessimism of the Intellect, Optimism of the Will.

Feynman and meMemories of Feynman. Labels: Feynman, path integrals.

Richard Feynman and the 19 year old me at my Caltech graduation:

Mama said knock you out (learning how to fight).

Many Worlds and quantum mechanics, a brief guide. Label: Many Worlds. Note, the usual quantum probabilities do not emerge naturally in this interpretation. See my papers On the Origin of Probability in Quantum Mechanics and The measure problem in no-collapse (many worlds) quantum mechanics.

Cognitive limitations: statistics , higher ed. Label: bounded rationality, human capital. Brainpower and globalization.

Expert predictions in soft subjects are unreliable. Intellectual honesty. Frauds! Label: expert predictions.

We can (crudely) measure cognitive ability using simple tests. (It is amazing to me that this is a controversial statement.) Randomly sampled eminent scientists have (very) high IQs, and given the observed stability of adult IQ the causality is clear: psychometrics works. The cult of genius? Income, Wealth, and IQOne hundred thousand brains. Bezos on the Big Brains. Label: psychometrics.

Historically isolated groups of humans cluster genetically according to geographical ancestry. Explained in pictures , words , more words.

I am skeptical of all but the weakest claims of market efficiency. My talk on the credit crisis. Venn diagram for economics.

Careers, advice to geeks: A tale of two geeks , success vs ability. Labels: careers , startups , entrepreneurs.

Net worth , life satisfaction , happiness , the gilded age.

What is the likely development path for China in the next decades? Sustainability of China growth , China development: how big is the middle class? , Back to the future , Shanghai from an Indian perspective.

That curious institution, Caltech. How did a 16 year old kid from Iowa end up there? (See memories of Feynman above.)

There are geniuses in the world. The cult of genius.

My lovely kids. Photos. Autobiographical.


Credentialism and elite careers , Defining meritelitism , brainpower

Recent videos (talks on genomics):

Talks (some with slides + video):

Cold Spring Harbor Laboratory
Berkeley Innovative Genomics Institute and OpenAI
Janelia Research Campus (HHMI)
Allen Institute (Seattle) meeting on Genetics of Complex Traits

Review article: On the genetic architecture of cognitive ability and other quantitative traits (2014)

I work on algorithms for phenotype prediction from genotype, using new methods from high dimensional statistics. My estimate is that prediction of complex traits such as height, cognitive ability, or highly polygenic disease conditions will require data sets of order one million individuals (i.e., to build a model which accounts for most of the genetic variance). Once these models are available, human reproduction (and evolution!) will be revolutionized.

These papers are somewhat technical:

This one is a bit less technical and gives a broader overview:

Cattle genomics (an existence proof):

These are for popular audiences (Nautilus Magazine):

As predicted, we now have good height predictors thanks to the 500k genome release of UK Biobank data: Scientists of Stature

Genomic predictors for common disease risk, constructed via machine learning on hundreds of thousands of genotypes. The predictors use anywhere from a few tens (e.g., 20 or 50) to thousands of SNPs to compute the risk PGS (Poly-Genic Score) for conditions such as diabetes, breast cancer, heart attack, and more: Genomic Prediction of Complex Disease Risk.

The Economist on polygenic risk scores (2019).

Detailed analysis of genetic architectures of disease risk predictors. Implications for pleiotropy.

Sibling validation of genomic predictors.

Links to recent papers from my group:


Ian Smith said...

"Randomly sampled eminent scientists have (very) high IQs"

Like Feynman? I read this. It is difficult to believe you fell for this BS. Even if the made up tests were valid the stability of adult IQ for the population at large does not mean stability of IQ for those who make a living using their intelligence.


SAT: 1560
GRE: 800, 800, 800
GMAT: 770

Steve Hsu said...

You make a valid point, but I believe it is answered in the Terman data. Adult IQs were not appreciably higher (relative to childhood scores) in the Terman subset that pursued challenging careers than in the subset that did not. Of course this point deserves more scrutiny -- the SMPY study could address it using their older cohorts.

Donald Pretari said...


For some reason, I have to tell you my Cal Tech story. I have always been a humanities type. In my high school in a small town in the Central Valley of Cal, the hardest and best class was chemistry. Everybody took it their senior year. Since I didn't care how I did, I took it as a junior with a female friend of mine who was our junior year science wiz, and the daughter of one of our science teachers. I also took physics that year.

Incredibly, in both the physics and chemistry class, I not only got A's, but the teachers told me that I had done the best in the class. The senior class included three people who went to Stanford, one as a chemistry major. My teacher also told me that I had done better than a girl in the previous year's course who went to Cal Tech.

I think that the reason this class was so excellent was that the teacher was a retired Army colonel who'd worked as a chemist in the army, and had undergraduate and graduate degrees in physics from Cal. Anyway, based on his statement, I decided to take the AP tests in chemistry, English, and history, very early in my senior year. I did good in history and English, but great in Chemistry. I think that it was the 98th or 99th percentile. I actually thought that it was a mistake. When I showed it to my chemistry teacher, he was quite pleased. So pleased, in fact, that he put my name in for a scholarship without telling me.

So, in the Fall of my senior year, I received a letter from an oil company or oil executive in Cal that informed me that I had won a very selective and lucrative scholarship for college. I was ecstatic, until I read the conditions. I had to major in chemistry or chemical engineering, and attend either Cal Tech or Harvey Mudd.

Now, I'd taken the SAT as a junior because a good friend of mine was being courted by colleges already. He became an NFL running back. I scored 600 in Math, which I considered fine. In other words, given my score and mediocre math grades, I wasn't going to Cal Tech or Harvey Mudd.

One final point. In the year that I graduated from high school, a Cal Tech grad from my high school won the Nobel Prize in Physics.

Don the libertarian Democrat

Ian Smith said...

Another thing Steve. The stability of adult IQ does not mean the stability of very high IQs or very low. There should be much greater variability in the IQ (in terms of IQ points) on retest at the extremes. A change from 130 to 145 is much more likely than from 100 to 115.

Unknown said...

> There should be much greater variability in the IQ (in terms of IQ points) on retest at the extremes.

"Greater", yes. "Much greater" - I think not.

As I recall, the mean change upon retest is only four points. Your own scores seem to correspond to something a little above IQ 155. The mean for highly accomplished scientists in the data cited by Steve is also in this approximate neighborhood for math and verbal, quite distant from the mean for all scientists. 4/55 ~= 7.3%, leaving a huge signal/noise ratio.

I'm not familiar with the GMAT, but notice that your own extremely unusual SAT concords with your GRE. According to you this should be rather unlikely. (Granted, there is some loss of information since you maxed out the GRE; we don't know for sure that you wouldnt've got 900 900 900 were that possible.)

My total SAT = my total GRE and they were quite far from the mean.

No effort to alter IQ has ever altered adult IQ compared to a control group. (The IQ of subadults is a bit more malleable, and, notably, less correlated to that of the individual's parents, siblings, identical twin, etc.)

There simply isn't a single fact, to my knowledge, that really tallies with a robust environmentalist position, and I would in fact be grateful if you could point one out, because I am always ready to give up my views. The closest thing to that would surely be the Flynn effect - it's arguably at least something of a fig leaf for the whole business, but I don't think it really tallies with robust environmentalism as it's commonly understood. It's mostly just kind of strange.

Ian Smith said...

A possible example: The difference between the mean IQ of British South Africans and that of Afrikaners. I don't remeber the source.

My scores mean nothing. I prepped.

Unknown said...

What non-anecdotal evidence do you have that prepping substantially succeeds.

Ian Smith said...

Is there any non-anecdotal evidence that it doesn't?

Unknown said...

I tried[SAT preparation]. Here are the top hits.

hit #1

"Special preparation resulted in an overall difference of eight points [...] between the treatment and control groups, corresponding to one additional correct item and stemming primarily from performance on analogy and antonym items."

hit #2 irrelevant

hit #3 you can read the first page for free - it's informative and rather concordant with hit 1:

#4 a cite only

#5 irrelevant

#6 Some stats jocks looking for something on which to test out their sophisticated observational study method. That's fine, but it's irrelevant to resolving the actual question, considering that randomized interventional studies on the question exist. Presumably the intended "real" use of this method is the examination questions where randomized study isn't possible.

Unknown said...

ahem, examination OF questions where randomized study isn't possible.

Anonymous said...


Two things

1) I’d like your permission to (re)print your article on ‘Torchwood’for our website

2) I was hoping we could use your ‘scribing’ talent for our website.

The Best Shows Youre Not Watching (dot) com [all one word]
‘Torchwood’one of our featured shows. We’re hoping to round up a few people who can occasionally contribute perspective (via an article/blog) on the shows – maybe a recent episode, future direction, plot shortcomings etc.

What’s in it for you?
Primarily a larger audience back channeled to your blog. We don’t pay but the site has a lot of promise and we're pretty excited about getting it off the ground. Let me know what you think.

Blog Archive