Toban Wiebe

A short essay on one of my favorite stats fun facts: a high-dimensional isotropic Gaussian puts almost all of its probability mass in a thin spherical shell centered on the mean, even though its density is highest at the mean itself.

My favorite stats fun fact is that high-dimensional isotropic Gaussian distributions have almost all of their probability mass concentrated in a thin spherical shell centered on the mean, like a soap bubble, even though their density is bell-shaped and ma…

Evaporative cooling is tempting to explain as a selection effect.

Histograms are usually the first plot people reach for when they want to look at a distribution.

I just published a new essay: Use eCDFs instead of histograms.

Data scientists worry too much about class imbalance.

I just launched What’s in a Name?

I published a new essay: Stop worrying about class imbalance.

Suppose you’re interested in knowing the chances of an event XXX occuring (e.g., X=X =X= “a nuclear strike over any populated area in the year 2019”).

Suppose you’re interested in knowing the chances of an event XXX occuring (e.g., X=X =X= “a nuclear strike over any populated area in the year 2019”).

Toban Wiebe is an Insight Data Science Fellow in Silicon Valley.

Anaconda is a very handy Python distribution that bundles a slew of scientific packages along with the handy conda package manager, allowing you to easily update your packages to the latest versions.

Update: with the release of Julia 1.0, the package manager, Pkg, was completely rewritten, though it remains quite similar.

Last week, GitHub Pages upgraded to Jekyll 3.0.

I’m a huge fan of Gmail’s vim-inspired keyboard shortcuts.

I somehow ended up reading this long essay by economist John Cochrane on growth-oriented policy.

I somehow ended up reading this long essay by economist John Cochrane on growth-oriented policy.

In the spirit of blogging like a hacker, I am now taking notes like a hacker, thanks to a system I pieced together using Markdown and Dropbox.

So I decided to move from Wordpress to Jekyll.

[...] there is a kind of puritanical view that everything relating to nicotine is bad and harmful and should be stamped on.

[...] there is a kind of puritanical view that everything relating to nicotine is bad and harmful and should be stamped on.

I've never felt comfortable with the logical positivists' "science is prediction" characterization, primarily because it neglects what I intuitively think of as the heart of science: explanation.

I've never felt comfortable with the logical positivists' "science is prediction" characterization, primarily because it neglects what I intuitively think of as the heart of science: explanation.