Unrelated to all that, 9/5 edition


FEINSTEIN: Hey, there’s something else you should know, Johnson. He’s not just with the graduate program. He’s an MD/PhD student.

JOHNSON: his mouth hangs open in rage He’s with the medical school? There’s no way I’m working with him now! We’ve been fighting with them over jurisdiction on this research project for months now. Remember that axon-pathfinding paper we almost had published before it was stolen by those patient-coddlers? We don’t need a future stethoscope-pusher hanging around our lab.

The phone rings, and FEINSTEIN answers it.

FEINSTEIN: Yeah? Really? Glutamatergic and GABAergic neurotransmitter vesicles? All right. I’ve got the perfect guys for the job. [He hangs up phone.]

FEINSTEIN: My inside man over at Nature Neuroscience tells me that those pseudo-scientists over at that Hopkins lab are trying to scoop us on your paper, Johnson. My pal is going to work over their submitted manuscript with some nasty reviews, but I’m going to need you two to get down to your bench and get me some electron microscopy images so we take ’em out.

How to Get Into an Ivy League College—Guaranteed

Ma’s algorithm, for example, predicts that a U.S.-born high school senior with a 3.8 GPA, an SAT score of 2,000 (out of 2,400), moderate leadership credentials, and 800 hours of extracurricular activities, has a 20.4 percent chance of admission to New York University and a 28.1 percent shot at the University of Southern California. Those odds determine the fee ThinkTank charges that student for its guaranteed consulting package: $25,931 to apply to NYU and $18,826 for USC. “Of course we set limits on who we’ll guarantee,” says Ma. “We don’t want to make this a casino game.”

Accidentally Turing-Complete

C++ Templates
Although they were initially not supposed to, C++ templates are Turing-complete. For proof look at this paper (pdf).

Magic: The Gathering
Magic is a card game. Apparently, the rules are complex enough to reach Turing-completeness.

The venerable mail server is known for its arcane configuration. Turns out the configuration is also turing complete.

Where America’s 46 million immigrants come from


(more visualizations here)

Metabolic costs and evolutionary implications of human brain development

The metabolic costs of brain development are thought to explain the evolution of humans’ exceptionally slow and protracted childhood growth; however, the costs of the human brain during development are unknown. We used existing PET and MRI data to calculate brain glucose use from birth to adulthood. We find that the brain’s metabolic requirements peak in childhood, when it uses glucose at a rate equivalent to 66% of the body’s resting metabolism and 43% of the body’s daily energy requirement, and that brain glucose demand relates inversely to body growth from infancy to puberty. Our findings support the hypothesis that the unusually high costs of human brain development require a compensatory slowing of childhood body growth.


Synapses: claymation style

Shenzhen trip report – visiting the world’s manufacturing ecosystem

The market was several large city blocks full of 5-10 story buildings with stalls packed into each floor. Each building had a theme or themes ranging from LEDs to cellphone hacking and repair. I realize it’s cliché to say this, but it REALLY felt like blade runner in a way that even Akihabara never did. I think it had a lot to do with the fact that many of vendors were selling to factories so were focused on wholesale and not retail and the volumes were huge and the interfaces were rough.

iPhone home buttons, wifi chipsets, Samsung screens, Nokia motherboards, everything. bunnie pointed to a bag of chips that he said would have a street value of $50,000 in the US selling for about $500. These chips were sold, not individually, but by the pound. Who buys chips by the pound? Small factories that make all of the cellphones that we all buy “new” will often be short on parts and they will run to the market to buy bags of that part so that they can keep the line running. It’s very likely that the “new” phone that you just bought from ATT has “recycled” Shenzhen parts somewhere inside.

A visual proof that neural nets can compute any function

This result tells us that neural networks have a kind of universality. No matter what function we want to compute, we know that there is a neural network which can do the job.

What’s more, this universality theorem holds even if we restrict our networks to have just a single layer intermediate between the input and the output neurons – a so-called single hidden layer. So even very simple network architectures can be extremely powerful.

The universality theorem is well known by people who use neural networks. But why it’s true is not so widely understood. Most of the explanations available are quite technical. If you’re a mathematician the argument is not difficult to follow, but it’s not so easy for most people. That’s a pity, since the underlying reasons for universality are simple and beautiful.

Inside Google’s Secret Drone-Delivery Program

Imagine all the possible ways one might get something from high in the air down to the ground. How about a tiny parachute à la The Hunger Games? Roy’s team tried it. There was too much wind interference and they struggled with accuracy. How about literally firing them down, a ballistic approach? “We contemplated this,” Roy said. “And then Sergey walked out from under a balcony and we almost hit him in a drop test.” After that, they moved on.

Who’s In The Office? The American Workday In One Graph


From the sestra-tumblr


One thought on “Unrelated to all that, 9/5 edition

  1. Nice collection of links this week. I like the little summaries, would like it even more if you offered some of your analysis/thoughts on them.

    It is very easy to jump into being Turing complete, one of the proofs that Magic is Turing Complete was generated on cstheory SE and garnered some excitement. A more subtle (reseach) issue now-a-days is to limit the power of your programming languages so that they aren’t Turing Complete but prevent the programmer from making stupid mistakes. See here for programming languages, and here for database query languages.

    It’s very likely that the “new” phone that you just bought from ATT has “recycled” Shenzhen parts somewhere inside.

    I think that sentence is intended as negative, but it really shouldn’t. It would be amazing if we recycled technology parts more, the fetishization of ‘new’ is completely unacceptable!

    The universality theorem is well known by people who use neural networks.

    I am often saddened by how often this sentiment is proven false. Some people in modeling (my experience is usually with people from computational psychology) seem to get so excited about their neural net learning a simple small-input problem from a huge amount of data.

    Less troll-y: I think focusing on the computability aspects can sometimes be a bit misleading; computable is simply too loose a consideration. Sometimes it is good (but much harder) to focus on the computational complexity aspects of neural nets, here you can start to separate different depths of nets a bit more and restrict the class of functions that are efficiently learnable (i.e. not requiring an super-polynomial number of hidden units, or a super-polynomial long training time).

    Who’s In The Office? The American Workday In One Graph

    I am pretty sure that first bump is just people browsing reddit.

Leave a Reply to Artem Kaznatcheev Cancel reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s