shas3

I think the diminishing returns is for the present and from the perspective of comforts of the present. From a savings perspective, there's no way more money in my retirement savings, kids' college funds, or for that $10k vacation won't make me happy. Who knows whether projections of 7% compounding rate of returns holds up for our generation's future!

Are there any features of brain-related signals that are biometric in nature?!

Big ships are slow to turn. One systemic cause based on my experience in old-big companies is that if profit margins on some products are high, mid and some high level managers maximize that product (e.g. roll-to-roll manufactured films in Kodak). With such optimization, you end up at "local-maxima". This continues until you have an "iceberg right ahead" moment when profits start to spiral downward. Depending on the pace of growth of the technology/sales of your competitor(s), this descent is rapid and irreversible.

We also need to hear from an important segment of the math population: those who burned out or dropped out for reasons related to math perceptions engendered by the math-competition culture. We all know people who did great at math or science competitions in high school but just disappeared from the scene after that. One may say that there are other underlying causes that lead them to not live up to some hypothetical promises, but I strongly feel that success and failure at math competitions can be a cause in itself. A lot of space is dedicated to how math olympiad triple-gold medalists went on to become great mathematicians. We also hear about those like Grothendieck and Hardy who weren't big in the competition circuit.

But missing are the stories of those who didn't make it big in spite of great competition performance, and those who fell out of math because of failing at math competitions.

In India, for example, competition math is everything at the high school level. This is because competitive exams like the famed IIT JEE, etc. are essentially variations on the competitive math theme. A few serious math enthusiasts do take up broader math-specific exams for math institutes, but those numbers are minuscule. The worst affected in my experience, are the talented and the enthusiastic who were discouraged and/or dropped out altogether because of failing at optimizing their skills and learning for competitions and similar exams.

It is generally true. Less 'fancy' work, you get less money, irrespective of how critical your code/work is. It is an irrational element that is not necessarily driven by supply-demand dynamics.

If you are interested in a longer account, I highly recommend mathematician/popularizer Marcus DuSautoy's Music of the Primes. https://en.m.wikipedia.org/wiki/The_Music_of_the_Primes

But where's the link to centrophenoxine and racetams? Your link only talks about anticholinergic drugs. I have seen such results about benzodiazepine http://www.health.harvard.edu/blog/benzodiazepine-use-may-ra...

When you are chemically toying with your brain, it is good to know if there are any adverse long terms effects. Risk vs. reward. Surely some non chemical ways of modifying the your brain can also have adverse effects: like eroding concentration due to constant multitasking.

Hope someone writes a similar letter about nootropics. There's little evidence about their long term effects.

This is pretty standard in old industrial companies.

Walmart Labs is basically a Bay Area software/analytics start up- Kosmix that was acquired by Walmart. Hardly surprising they do this.

On another note: a quirky thing I encountered on Twitter was Target hiring category theorists. https://twitter.com/sigfpe/status/734129734569709568

That's why you have Marshall Project. https://www.themarshallproject.org/#.t1nzcmhDj

Sure, if you have money, it helps in that you are likely to be aware of the Marshall Project, able to contact them, etc. But other than that, it is accessible to everyone.

On the academic-industry/start up side, I know of one such device which does this on a sphere: http://visisonics.com/visisonics-audiovisual-camera/

It works quite well in my experience with it. The spherical geometry enables a 3D capture and visualization.

As someone who has only visited SV, I wonder: In order to avoid the crazy living expenses in SV and SF, why don't more people start companies within 100 miles of SV, in cheaper places like San Ramon/Castro Valley, Livermore, Sacramento, Pleasanton, Modesto, etc.? That way you have reasonable proximity without having to suffer the living quality issues associated with SV and SF.

There is precedent for these types of acoustic attacks. One interesting paper from 2014 that comes to mind is this: https://www.tau.ac.il/~tromer/acoustic/

Basically, the idea is that even though clocks run at GHz (which in acoustics, would be impractical ultrasound), modular arithmetic (exponentiation) takes ~milliseconds to run, which means they produce acoustic signatures in the kHz range, which travels easily (and omnidirectionally from the kind of small aperture that the motherboard represents) in air and can be acquired with cheap microphones.

Some chemical companies still have scientific glassblowing departments.

I think both can be right. Ng and LeCun are talking about the real near future. Bostrom always came across as a speculative SF + philosophy kind of guy. Are there any specific critiques of his (and Yudkowsky/MIRI, Musk, etc.) arguments? I think the two claims are plausible:

1. AGI is imminent: 50 years or 500 years or more from now. This is not too unlikely given that the brain is just an information processing system. Emulating it is likely to happen sometime in the future.

2. Such an AGI will be all powerful because it is not limited by human flaws. Trivial or not, we will have to program it with "thou shalt not kill" type values.

I think there is a categorization error here. Deep architectures are possible for many classes of learning algorithms: deep Bayesian nets, restricted Boltzman machines, graphical models, multilayer perceptrons, etc. Convolutional neural nets and LTSM are just two other types, and they exploit spatial and temporal structure explicitly.

Larger problem: mnist, images, and speech are probably low hanging fruits with rich features, large datasets, balanced datasets, etc. for which off the shelf packages work very well today.

It almost sounds like a farce: Clinton gets indicted or loses face due to incompetence relating to the e-mail server scandal, and in her place is elected The Donald, a man who has committed worse transgressions and has the potential to do even worse things when given power.

Frankly, she doesn't sound any more incompetent than even a typical old company c. 2010: think Target, Sony, etc. It would be sad, given that her opponent will likely be Trump, if this scandal sinks her candidacy.

I know people who work as "fixers" specializing in navigating such government restrictions and bureaucracy for local and foreign companies in India.

It is more likely that the real reason for Apple's rejection is the whim of some high level bureaucrat ("IAS officer"). It may not even be corruption, even righteous bureaucrats are often stupid. The amount of red-tape in India is astounding.

Foreign students are 1.13 million [1] and the sum total of all students in the US is around 20 million ( age < 25 + age > 25) [2]. How much can they really contribute marginally to tuition inflation given that they are only 5-10% of the population? If foreign workers are taking American jobs, Chinese factories are taking over manufacturing, and international students are leading to tuition inflation, in general, US seems to be coping rather poorly with the effects of globalization. More likely, internationals are like Don Quixote's windmills.

Edit: forgot references

[1] http://www.wsj.com/articles/international-students-stream-in...

[2] http://nces.ed.gov/fastfacts/display.asp?id=98

> Some of people are calling themselves "Data Scientists" who don't know the difference between σ and MAD?

Uh, that's just terminology. Data science is quite diverse and people have different backgrounds and work in different domains. Many mathematically/CS/EE-trained data scientists probably just look at the whole thing in terms of norms without explicitly using the term, MAD.

But this is bound to happen whenever a region reaches a level of success. There will be incumbents who are enjoying their cushy positions which they rode into when the times were greater than ever (e.g. Boston Brahmins, people in 'company towns', early mom-and-pop/neighbor/Stanford-prof/etc. investors in SV companies like Google, FB, etc.). The children and proteges of the first couple of successful generations of Bay Area are now working age or older. They have their own networks formed in high school, neighborhoods, parents' companies and so on. I think this kind of establishment of an establishment starts once the first successful generation has children. One indicator of the 'establishment' is you have people using family connections- 'my dad's friend is the VP of blah blah,' etc.

Most post secondary math, especially proof-oriented abstract math- analysis, topology, abstract algebra, probability and measure- all the kind of stuff that Tao probably teaches- really cannot be visualized with modern technology and relies more on pen-on-paper/chalk-on-board walkthroughs. In that context, Tao is spot on that modern tech is of little use in exposition.

Yeah, I don't understand how Portland will be the next SF or even the next Seattle. SF and Seattle are years ahead in economic growth. Further, SF and Seattle have solid schools and universities. Portland is quite mediocre in that respect. The local universities in Portland and in wider Oregon are eminently mediocre.

I am reminded of an interesting observation by the mathematician Terence Tao about how our anonymity on the Internet and in a connected world is so fragile [1]. Basically, because there are only 3 billion internet users, every person can be uniquely identified by a 31 bit number. The uncovering of each bit gets one closer to the identity of the person. Seen in this light, one would expect metadata to uncover quite a few of the 31 bits!

[1] https://plus.google.com/+TerenceTao27/posts/8vmpA9fgRMq?iem=...

> The only mystery to me was why Intel never just sucked it up and made a low-margin x86 chip to compete directly with ARM.

High-margin addiction is typically deeply embedded in a company's culture. Every manager who is accountable for revenue is keenly aware of the incentives of toeing the high-margin-culture line. As a result, you have low-margin products and ideas killed off by mid-level managers. For big companies, there is no choice: you either get lucky with high-margin addiction, or you die.

Mine is but one perspective, as most comments everywhere are. One can also add Bloomfield to the list of neighborhoods with hipster-gentrification.

There's certainly a hipster bubble in Pittsburgh around Oakland, Shadyside, -W-e-s-t- East Liberty, Southside, and Squirrel Hill which are the neighborhoods around CMU and U Pitt. The rest of Pittsburgh and the suburban region is pretty rough, though. Personally, I loved North Hills, Mt. Lebanon, and Dormont. But the start-up quality is only present in the neighborhoods abutting CMU and U Pitt. As a non-white, we were sometimes bothered by the extreme whiteness and slight red-necky feel of non-East End neighborhoods.

Edit: as pointed out by comments below, East Liberty.

Just saying, working touch screens have been around on devices for 30 years or more. Most 'PDA's, including < $100 phonebooks had decent resistive touch screens as far back as the mid-90s. Also, as such, speech recognition too has been around for at least 25 years. What's new in today's smartphones is the cloud+deep learning driven high-accuracy. You could probably have easily found a handheld with touchscreen and speech recognition at Circuit City in 1999.