Terminological differences
- “One source of confusion is the use of new terminology in ML for concepts that have well-established labels in the older literatures. In the context of a regression model, the sample usedto estimate the parameters is often referred to as the training sample. Instead of the model beingestimated, it is being trained. Regressors, covariates, or predictors are referred to as features. Re-gression parameters are sometimes referred to as weights. Prediction problems are divided intosupervised learning problems, where we observe both the predictors (features)Xiand the outcomeYi, and unsupervised learning problems, where we only observe theXiand try to group them intoclusters or otherwise estimate their joint distribution. Unordered discrete response problems aregenerally referred to as classification problems.” (???)