Abstract

We establish theoretical results concerning local optima of regularized M estimators, where both loss and penalty functions are allowed to be nonconvex. Our results show that as long as the loss satisfies restricted strong convexity and the penalty satisfies suitable regularity conditions, any local optimum of the composite objective lies within statistical precision of the true parameter vector. Our theory covers a broad class of nonconvex objective functions, including corrected versions of the Lasso for errors-in-variables linear models and regression in generalized linear models using nonconvex regularizers such as SCAD and MCP. On the optimization side, we show that a simple adaptation of composite gradient descent may be used to compute a global optimum up to the statistical precision  in iterations, the fastest possible rate for any first-order method. We provide simulations to illustrate the sharpness of our theoretical predictions.