Sparse and Smooth Signal Estimation: Convexification of L0 Formulations

11/06/2018
by   Alper Atamturk, et al.
0

Signal estimation problems with smoothness and sparsity priors can be naturally modeled as quadratic optimization with ℓ_0-"norm" constraints. Since such problems are non-convex and hard-to-solve, the standard approach is, instead, to tackle their convex surrogates based on ℓ_1-norm relaxations. In this paper, we propose new iterative conic quadratic relaxations that exploit not only the ℓ_0-"norm" terms, but also the fitness and smoothness functions. The iterative convexification approach substantially closes the gap between the ℓ_0-"norm" and its ℓ_1 surrogate. Experiments using an off-the-shelf conic quadratic solver on synthetic as well as real datasets indicate that the proposed iterative convex relaxations lead to significantly better estimators than ℓ_1-norm while preserving the computational efficiency. In addition, the parameters of the model and the resulting estimators are easily interpretable.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset