While Model Trains

Read data blog posts.
Carefully handpicked.
Presented 3 at a time.

Recommendations

Simon Hørup Eskildsen

Cool explanation of computational aspects of recommendations avoiding to "get lost in the weeds of linear algebra".

Read it!

Why Correlation Usually ≠ Causation

Gwern

"Despite this admonition, people are overconfident in claiming correlations to support favored causal interpretations and are surprised by the results of randomized experiments, suggesting that they are biased & systematically underestimate the prevalence of confounds / common-causation."

Read it!

Finding bad flamingo drawings with recurrent neural networks

Colin Morris

Using Sketch-RNN as a probability estimator to identify the worst sketches of flamingos in 'Quick, Draw!'.

Read it!