A0386
Title: Fused lasso nearly-isotonic signal approximation in general dimensions
Authors: Vladimir Pastukhov - Chalmers University of Technology (Sweden) [presenting]
Abstract: The purpose is to introduce and study fused lasso nearly-isotonic signal approximation, which is a combination of fused lasso and generalized nearly-isotonic regression. We show how these three estimators relate to each other and derive solutions to the general problem. Our estimator is computationally feasible and provides a trade-off between monotonicity, block sparsity and goodness-of-fit. Next, we prove that fusion and near-isotonisation in the one-dimensional case can be applied interchangeably. Also, we derive an unbiased estimator of the degrees of freedom of the estimator.
