L0 regularization is a sparsity-inducing regularization approach that penalizes the number of nonzero model weights or features. Because exact L0 optimization is difficult, practical systems often use approximations or alternative regularizers to encourage simpler models.
A type of regularization that penalizes the total number of nonzero weights in a model. For example, a model having 11 nonzero weights would be penalized more than a similar model having 10 nonzero weights. L~0~ regularization is sometimes called L0-norm regularization. L~0~ regularization is generally impractical in large models because L~0~ regularization turns training into a convex optimization problem.