"That may be true," [the Dodecahedron] acknowledged, "but it's completely accurate, and as long as the answer is right, who cares if the question is wrong?"
24 July 2015
20 April 2015
Heisenberg, Schrödinger, and Ohm are together in a car, driving down the road.
They get pulled over. Heisenberg is driving and the cop asks him "Do you know how fast you were going?" "No, but I know exactly where I am" Heisenberg replies. The cop says "You were doing 55 in a 35." Heisenberg throws up his hands and shouts "Great! Now I'm lost!"
The cop thinks this is suspicious and orders him to pop open the trunk. He checks it out and says "Do you know you have a dead cat back here?" "We do now, asshole!" shouts Schrödinger.
The cop moves to arrest them. Ohm resists.
24 March 2015
What a perfect time to learn this lesson but during March Madness, via a winning prediction algorithm that does so well because it figured out what were the best data to use. This is called "variable selection" in statistics, and we try to do it in an automated way, but often (as was the case here) it's really domain expertise that allows one to figuring out which data are the best for inference.
One of my favorite lists of problems with big data can be found here-- including the fact that "although big data is very good at detecting correlations, ... it never tells us which correlations are meaningful." Here is another nice article from last summer on the limitations of big data -- through ok cupid and facebook user experiments. And of course my earlier blog post that gives props to IEEE for discussing the same.
Every statistical inference procedure -- from simply calculating p-values, to predicting class labels with SVM, to estimating system dynamics with filters -- has assumptions that may or may not hold in practice. Understanding the implications of that is crucial to figuring out how to use big data.
08 January 2015
"The fatherhood bonus also dissipates when men become more involved at home. ... Men who decrease their work hours for family reasons suffer a 15.5 percent decline [in future salary], while women’s salaries decline by just 9.8 percent. In other words, having a family helps men in the workplace only if they submit to their traditional gender role."
Park was a clerk for Justice Ruth Bader Ginsburg. I really enjoyed the video interview with Justice Ginsburg. Here is an interesting excerpt:
Park: "You mentioned your work at the ACLU women's right's project. ... Many commenters, the first thing they talk about was the tactical brilliance of bringing these cases where men were complaining about gender distinctions in the law that they believed harmed them. Can you talk a little bit about that strategic choice? Was it a strategic choice? Do you think the course of jurisprudence would have been different had you only brought cases with a female plaintiff?"
Ginsburg: "I have been complemented for mapping out a strategy, in truth it's the cases that came trooping into the ACLU.
"The turning point case, Reed v Reed, was a woman plaintiff. A law that discriminated blatantly against a woman, said 'As between persons equally entitled to administer a decedent's estate, males must be preferred to females.' Great first case.
"But then, let's talk about Stephen Wiesenfeld case, because that's a perfect example of what's wrong with drawing rigid lines based on gender. So Stephen was married to a teacher who had a very healthy pregnancy. She was in the classroom until the ninth month. She went to the hospital for the birth of the child. The doctor came out and told Stephen you have a healthy baby boy, but your wife died of an embolism. So he at that moment decided that he would not work full time until the baby Jason was going to school full time. And he had heard that the social security system has benefits for a sole surviving parent with a child under the age of 12 in his care. He applied for that benefit and was told, well we're sorry Mr. Wiesenfeld, these are mothers' benefits. Why mothers? Because mothers take care of children.
"Where did this discrimination begin? It began with the woman as wage earner. She paid the same social security tax as a man paid, but her family did not get the same protection that a man's family would. That was the majority view of the court, that it was really discrimination against the woman. And then one, who later became my chief, he was then Justice Renquist, said, this is utterly irrational from the point of the baby. Why should the baby have the opportunity for the care of a sole surviving parent if that parent is female, but not if the parent is male? So that was my example of how these rigid gender lines in the law hurt everybody."
08 December 2014
20 November 2014
Have you heard yet that in 2010 Barbie was an incompetent "computer engineer" in a kid's book by Mattel? Before you think to yourself, "Come on, any book about Barbie being a computer engineer must be good!" -- please, go look at the pictures of the book. (and actually it seems, she was a graphic designer.)
That excellent critique went viral this week, and Mattel has pulled that book from amazon.com. That's awesome.
Even more awesome is this wherein you can remake Barbie into a real computer programmer making an awesome game.
* Hopefully we can think of this as "throwback Thursday."