From Event: SPIE Optical Engineering + Applications, 2017
This paper studies least-square regression penalized with partly smooth convex regularizers. This class of penalty functions is very large and versatile, and allows to promote solutions conforming to some notion of low-complexity. Indeed, such penalties/regularizers force the corresponding solutions to belong to a low-dimensional manifold (the so-called model) which remains stable when the penalty function undergoes small perturbations. Such a good sensitivity property is crucial to make the underlying low-complexity (manifold) model robust to small noise. In a deterministic setting, we show that a generalized "irrepresentable condition" implies stable model selection under small noise perturbations in the observations and the design matrix, when the regularization parameter is tuned proportionally to the noise level. We also prove that this condition is almost necessary for stable model recovery. We then turn to the random setting where the design matrix and the noise are random, and the number of observations grows large. We show that under our generalized "irrepresentable condition", and a proper scaling of the regularization parameter, the regularized estimator is model consistent. In plain words, with a probability tending to one as the number of measurements tends to infinity, the regularized estimator belongs to the correct low-dimensional model manifold. This work unifies and generalizes a large body of literature, where model consistency was known to hold, for instance for the Lasso, group Lasso, total variation (fused Lasso) and nuclear/trace norm regularizers. We show that under the deterministic model selection conditions, the forward-backward proximal splitting algorithm used to solve the penalized least-square regression problem, is guaranteed to identify the model manifold after a finite number of iterations. This manuscript is an excerpt from the work published in .
Samuel Vaiter, Gabriel Peyré, and Jalal Fadili, "Recovery guarantees for low complexity models (Conference Presentation)," Proc. SPIE 10394, Wavelets and Sparsity XVII, 103941V (Presented at SPIE Optical Engineering + Applications: August 09, 2017; Published: 21 September 2017); https://doi.org/10.1117/12.2274752.5583610419001.
Conference Presentations are recordings of oral presentations given at SPIE conferences and published as part of the conference proceedings. They include the speaker's narration along with a video recording of the presentation slides and animations. Many conference presentations also include full-text papers. Search and browse our growing collection of more than 12,000 conference presentations, including many plenary and keynote presentations.