5,738 followers
70 followers
516 followers
"Explainable ML–using a black box and explaining it afterwards Interpretable ML–using a model that is not black box" My take: this is a false dichotomy. How can we *explain* cognitive flexibility of human problem solving, which is trivially interpretable?
2,251 followers
@omaclaren @dan_p_simpson @betanalpha i think so too -- "interpretable" here means transparent the way a decision tree is transparent (or Rudin's optimized list of rules, https://t.co/w78a8XLpEN) , as opposed to, say, deep neural nets (or for Bayesians, pe