On the total variation regularized estimator over the branched path graph

06/04/2018
by   Francesco Ortelli, et al.
0

We generalize to tree graphs obtained by connecting path graphs an oracle result obtained for the Fused Lasso over the path graph. Moreover we show that it is possible to substitute in the oracle inequality the minimum distance between jumps by the harmonic mean of the distances between jumps. In doing so we prove a lower bound on the compatibility constant for the total variation penalty. Our analysis leverages insights obtained for the path graph with one branch to understand the case of more general tree graphs. As a side result, we get insights into the irrepresentable condition for the branched path graph.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
06/04/2018

On the total variation regularized estimator over a class of tree graphs

We generalize to tree graphs obtained by connecting path graphs an oracl...
research
02/28/2019

Oracle inequalities for square root analysis estimators with application to total variation penalties

We study the analysis estimator directly, without any step through a syn...
research
01/14/2020

Graph-Fused Multivariate Regression via Total Variation Regularization

In this paper, we propose the Graph-Fused Multivariate Regression (GFMR)...
research
08/06/2019

An Algorithm for Graph-Fused Lasso Based on Graph Decomposition

This work proposes a new algorithm for solving the graph-fused lasso (GF...
research
07/26/2022

Variance estimation in graphs with the fused lasso

We study the problem of variance estimation in general graph-structured ...
research
07/03/2016

Variational limits of k-NN graph based functionals on data clouds

We consider i.i.d. samples x_1, ..., x_n from a measure ν with density s...
research
04/24/2019

Prediction bounds for (higher order) total variationregularized least squares

We establish oracle inequalities for the least squares estimator f̂ with...

Please sign up or login with your details

Forgot password? Click here to reset