Tweeted By @dohmatobelvis

The bridge between information theory and machine learning is called: Fano theorem. It begets powerful tools for computing lower bounds for minimax error rates in learning, only using metric quantities (covering number, etc.). Nice tutorial by John Duchi https://t.co/Wk1MrDv1IV pic.twitter.com/bLh0a26BvG

— Elvis Dohmatob (@dohmatobelvis) July 23, 2018