Solving Equations of Random Convex Functions via Anchored Regression

02/17/2017
by   Sohail Bahmani, et al.
0

We consider the question of estimating a solution to a system of equations that involve convex nonlinearities, a problem that is common in machine learning and signal processing. Because of these nonlinearities, conventional estimators based on empirical risk minimization generally involve solving a non-convex optimization program. We propose anchored regression, a new approach based on convex programming that amounts to maximizing a linear functional (perhaps augmented by a regularizer) over a convex set. The proposed convex program is formulated in the natural space of the problem, and avoids the introduction of auxiliary variables, making it computationally favorable. Working in the native space also provides great flexibility as structural priors (e.g., sparsity) can be seamlessly incorporated. For our analysis, we model the equations as being drawn from a fixed set according to a probability law. Our main results provide guarantees on the accuracy of the estimator in terms of the number of equations we are solving, the amount of noise present, a measure of statistical complexity of the random equations, and the geometry of the regularizer at the true solution. We also provide recipes for constructing the anchor vector (that determines the linear functional to maximize) directly from the observed data.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/13/2016

A Grothendieck-type inequality for local maxima

A large number of problems in optimization, machine learning, signal pro...
research
05/06/2021

Inverse Scale Space Iterations for Non-Convex Variational Problems Using Functional Lifting

Non-linear filtering approaches allow to obtain decompositions of images...
research
01/29/2019

Rank-one Convexification for Sparse Regression

Sparse regression models are increasingly prevalent due to their ease of...
research
06/19/2018

Estimation from Non-Linear Observations via Convex Programming with Application to Bilinear Regression

We propose a computationally efficient estimator, formulated as a convex...
research
10/27/2021

Spectrahedral Regression

Convex regression is the problem of fitting a convex function to a data ...
research
09/12/2016

Adaptive matching pursuit for sparse signal recovery

Spike and Slab priors have been of much recent interest in signal proces...
research
07/23/2020

On Solving Word Equations via Program Transformation

The paper presents an experiment of solving word equations via specializ...

Please sign up or login with your details

Forgot password? Click here to reset