Sparsity by Redundancy: Solving L_1 with a Simple Reparametrization
share
We identify and prove a general principle: L_1 sparsity can be achieved using a redundant parametrization plus L_2 penalty. Our results lead to a simple algorithm, spred, that seamlessly integrates L_1 regularization into any modern deep learning framework. Practically, we demonstrate (1) the efficiency of spred in optimizing conventional tasks such as lasso and sparse coding, (2) benchmark our method for nonlinear feature selection of six gene selection tasks, and (3) illustrate the usage of the method for achieving structured and unstructured sparsity in deep learning in an end-to-end manner. Conceptually, our result bridges the gap in understanding the inductive bias of the redundant parametrization common in deep learning and conventional statistical learning.
READ FULL TEXT