Cortically Magnified

In a previous blog post, I reviewed Ellery Queen’s classic detective novel The Greek Coffin Mystery (1932), which manages to involve color blindness in its puzzles. But I am not done yet! I have a couple more. On one hand, it feels to me that color blindness is gimmicky as a plot device. A mystery writer must be quite desperate for new ideas if she or he has to turn to perceptual psychology (or any branch of specialized knowledge, for that matter).

I don’t think mystery novels by Ellery Queen are popular in western countries anymore, but they are still read in Asia. When I was a PhD student, every time I had to travel from my home country Taiwan to the USA, I would buy an Ellery Queen novel at the airport bookstore. This way, I could land in LAX with a solved mystery. Ellery Queen novels are substantial books with very complex plots - perfect for long flights because uninterrupted concentration is needed to tackle them.

I wrote this post primarily from the perspective of a beginner guitarist trying to solve a practical problem, but the way I approached it naturally landed on functional programming and literate programming. So, you might want to stay around even if you are not interested in playing the guitar. This post is mostly about my experience of learning a very interesting music markup language LilyPond, which has a symbiotic relationship with the GNU Guile Scheme.

The Simpson’s Paradox is one of the most well-known paradoxes in statistics. A quick google will find plenty of blog posts (many from the data science community) about this puzzling phenomenon. It is clearly a topic of real-world significance. There seem to be some important lessons that we are supposed to learn from it. But what are those lessons? Is it nothing more than a cautionary tale about how easy it is for data analyses to go wrong?

A recent paper published in PNAS titled “The Fermi-Dirac distribution provides a calibrated probabilistic output for binary classifiers” caught my attention, because it describes a surprising relationship between machine learning and quantum physics. In fact, surprising is an understatement. Mind-boggling is more like it. According to the analogy developed by the authors, positive samples in binary classification problems are like… fermions?! What?! I decided that I should try to understand the gist of this paper, at least to the extent that I can.