While Model Trains

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

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!

The dangers behind image resizing

Marco Venturelli

"It is essential to carefully choose the library you want to use to perform resize operations, particularly if you're going to deploy your ML solution."

Read it!

Deep Learning ideas that have stood the test of time

Denny Britz

An overview of deep learning breakthroughs that have demonstrated their value over time, from AlexNet in 2012 to GPT-3 in 2020.

Read it!