Double Descent in Risk Curves

A Brief Prehistory of Double Descent

In their thought-provoking paper [1], belkin et al.Illustrate and discuss the shape of risk curves in the context of modern high-complexity learners.These curves show the risk of a learner as a function of some (approximate) measure of its complexity.With increasing the risk initially decreases, attains a minimum, and then increases until equals where the training data is fitted perfectly.