August 22, 2011

Brains and computers

Posted in General at 15:12 by Tatyana Deryugina

From an interesting New Scientist article about a simulated rat that learns:

“The more fundamental issue is that brute-force computation cannot compare with the functioning of a real brain. Human intelligence arises not from logic, but from our ability to respond to ambiguity and adjust to rapidly changing situations. “The idea that human expertise can be formalised in logical rules turned out to be a fundamentally wrong assumption,” says Rolf Pfeifer, an AI researcher at the University of Zurich in Switzerland.”

Even more interesting:

“A brain consumes less power than a light bulb, and occupies less volume than a 2-litre bottle of soda,” says Dharmendra Modha, a computer scientist at IBM.

 

Advertisements

June 29, 2011

A note on polling

Posted in news, science at 10:57 by Tatyana Deryugina

I just saw a story claiming that TSA agents are developing cancer from working near the body scanner machines and that the TSA is covering it up. Personally, I do wonder about the cost/benefit of the scanners. Of course, cancer “clusters” could be random, and the article didn’t mention anything about statistical testing. It also made the following statement: “Of course, if TSA workers who are merely standing near the scanners are already developing cancer, frequent flyers are also putting themselves in harm’s way by standing directly inside the radiation-firing machines.”

That logic is clearly flawed. It’s not obvious how much exposure TSA workers get by standing next to the scanners. Of course, per unit of time, you get exposed to more radiation inside the scanner. But if you’re there for 40 hours a week, you might get a much larger dose of radiation than even the most active traveler. So this doesn’t prove the scanners are dangerous for travelers. However, we might care about TSA officials’ health too.

The website also had an amusing one-question survey  on the side: “Reports indicate naked body scanners are so dangerous, TSA agents are getting cancer from just being near them. If the TSA continues to use these demonstrably deadly devices, will you fly again?” (Yes/No)

Note the words “naked” and “demonstrably deadly” in there. Someone clearly didn’t read books on the proper way to do surveys or (more likely) decided to ignore good practice. I can just see the headlines: “75% of people say they will never fly again if the TSA continues to use body scanners”.

June 23, 2011

Pavlovian training for running

Posted in science, simply amusing at 20:04 by Tatyana Deryugina

There is direct evidence that endorphins are released following a period of strenuous exercise. What it takes for endorphins to be released is unclear (and probably varies from person to person), but the consensus seems to be that the body has to cross over some threshold of strain before endorphins are released. I’ve certainly experienced this myself (I think). The first two miles or so of my recent runs are usually pretty tedious and unpleasant.  I get tired and want to stop. Sometimes I get side pains. But then I start feeling better and am able to run another three miles without significantly slowing down. I’ve never gotten side pains during the second part of a long run (and it’s not because I selectively stop running). So I do think the time-delayed endorphin release is real.

Which brings me to the million dollar question – is there a way to get the endorphins to be released right before a run, to spare the body the pain of the pre-threshold strain?  Here’s my X million dollar idea (where X>=0): use classical conditioning, the idea behind Pavlov’s dog, to train the body to release endorphins. Have someone run for a while. Once the subject experiences an endorphin release, play a particular song. Eventually, the body will be trained to associate the song with releasing endorphins. Play the song right before you’re ready to run and voila! Instant endorphin-filled run!

Seriously, I wish someone would do this experiment and see if it works. I may try it on myself and report back.

June 22, 2011

Hayek and Anarchy

Posted in musings on economics at 16:12 by Tatyana Deryugina

Last month’s Journal of Economic Behavior & Organization has an interesting article titled “Hayekian anarchism”. I’m reposting the abstract below, but here’s the punchline: “Hayek should have been an anarchist.”

“Should law be provided centrally by the state or by some other means? Even relatively staunch advocates of competition such as Friedrich Hayek believe that the state must provide law centrally. This article asks whether Hayek’s theories about competition and the use of knowledge in society should lead one to support centrally provided law enforcement or competition in law. In writing about economics, Hayek famously described the competitive process of the market as a ‘discovery process.’ In writing about law, Hayek coincidentally referred to the role of the judge under the common law as ‘discovering’ the law in the expectations and conventions of people in a given society. We argue that this consistent usage was more than a mere semantic coincidence–that the two concepts of discovery are remarkably similar in Hayek’s thought and that his idea of economic discovery influenced his later ideas about legal discovery. Moreover, once this conceptual similarity is recognized, certain conclusions logically follow: namely, that just as economic discovery requires the competitive process of the market to provide information and feedback to correct errors, competition in the provision of legal services is essential to the judicial discovery in law. In fact, the English common law, from which Hayek drew his model of legal discovery, was itself a model of polycentric and competing sources of law throughout much of its history. We conclude that for the same reasons that made Hayek a champion of market competition over central planning of the economy, he should have also supported competition in legal services over monopolistic provision by the state–in short, Hayek should have been an anarchist.”

June 20, 2011

Can paid content compete with the internet?

Posted in General, news, science at 21:00 by Tatyana Deryugina

I’ve been a subscriber of Scientific American (paper, not online) for several years. For various reasons, I decided to stop subscribing to it and look for another source of detailed science news (one that went into more technical detail than the Economist’s Science and Technology, SciAm or Discover). After finding various potential candidates, I decided to square them against each other in an RSS Feed War. The best one gets a print subscription from me. After subscribing to the feeds, I realized that a paid subscription may prove redundant. The only way I see myself getting one is if there’s too much information coming through the feeds. Although I only subscribed today, I now have about 300 unread entries (each new feed shows the latest 10 entries as unread). Ironically, the benefit of paper magazines might be that it’s much more costly to produce than online text and thus has to contain the most relevant material. Not that this is impossible to do with online content, but it doesn’t look good so far.

Here are my candidates, in no particular order (feel free to suggest others):

Some of the candidate magazines/websites have topic feeds, so there’s more than one feed per “candidate”.

June 13, 2011

Sects of Islam

Posted in General at 17:13 by Tatyana Deryugina

I have to admit that although I first heard the terms “Sunni Muslims” and “Shia Muslims” years ago, I never bothered to figure out what the differences between the two sects of Islam were. Recently I finally sat down and looked it up. Turns out, the differences aren’t major. But the origins of the sects are pretty interesting – they arose over a disagreement about who should succeed Muhammad, someone elected by the the people or a blood/undemocratically chosen heir. You can find the details here. Here’s a nice chart with the major differences in beliefs between the two.

June 9, 2011

The global cooling “consensus” of the 1970’s

Posted in science at 08:34 by Tatyana Deryugina

For those who have heard that scientists in the 1970’s believed a new Ice Age was imminent, I recommend reading this article, which investigates what the true consensus was back then. To give away the punchline, the title of the article is “The Myth of the 1970s Global Cooling Scientific Consensus”. It’s non-technical and very well-written.

May 28, 2011

Moonwalking with Einstein, Part 3

Posted in books, simply amusing at 16:29 by Tatyana Deryugina

I finally finished reading “Moonwalking with Einstein” by Joshua Foer, where he documents his quest to win the US Memory Championship (even though the quest starts out with a less ambitious goal).  It was a very easy to read and contained a lot of interesting facts. For one, I had no idea that memorization was such a big part of life in ancient times. When you think about it, it makes sense – many people didn’t know how to read and those who did couldn’t afford to own every book. But I never really thought about it.

Joshua also explores the status of memorization in various spheres of life today, from the Memory Championships and those who participate in them, to scientists who study memory, to educators who debate its role in education. He describes several past and present memory savants and even accuses one of them of being a well-trained mnemonist. Overall, it’s a worthwhile and entertaining read.

Earlier, I wrote about using the memory techniques in this book to memorize all the countries in the world. I made it through all of Central/South America and 2/3 of Africa. The techniques work really well. But you have to be motivated to use them. And, as the author discusses in the book, we live in a world where memorization is mostly unnecessary. Although I may try to use it to memorize names of people I just met.

May 22, 2011

Poor Economics

Posted in books, musings on economics at 18:44 by Tatyana Deryugina

I just finished reading “Poor Economics” by Abhijit Banerjee and Esther Duflo, both MIT economics professors. The book was amazing and I highly recommend it. In summary, it is an excellent, evidence-based discourse about the behavior of the poor and the policies that work and don’t work to improve their lives. Abhijit and Esther cover how the poor make decision about how much to save, eat, and spend on their children’s education, why so many poor households run businesses but don’t become rich, and how political institutions can be improved.

Having never taken development economics, I learned a lot from the book. At the same time, it was written in everyday language and everything was incredibly clear and well-explained. Even though the authors clearly do have opinions, they arrive at most of them through examining the results of carefully designed research. They do their best to avoid generalization, exploring the complexity of the issues and dispelling the myths that the poor are somehow fundamentally different from the rest of us. Some of the findings were surprising and went against what I and others would have guessed (thus underscoring the need for rigorous research).

May 20, 2011

Verisimilitudes

Posted in verisimilitudes at 09:46 by Tatyana Deryugina

I was going to start an original series on “verisimilitudes”, ideas that appear true and persist in popular or even scientific beliefs, but are actually false. When I started researching a few myths, I realized that there are already many other people out there writing up myths. So I decided to start by sharing and summarizing existing webpages debunking myths. If I run out of good stuff (what’s the probability of that on the internet?), I’ll start doing more original research.

The myth of the day is: peeing on someone after they’ve been stung by a jellyfish helps. In fact, according to this article, it can even make the pain worse. I’ve never had this remedy applied to me, but I’ve definitely heard this advice and saw it applied in action.

Next page