Skip to content

demo for "A statistical mechanics approach to de-biasing and uncertainty estimation in LASSO for random measurements"

Notifications You must be signed in to change notification settings

takashi-takahashi/debiasing_lasso_demo

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 

Repository files navigation

A statistical mechanics approach to de-biasing and uncertainty estimation in LASSO for random measurements

image_of_debiased_estimator

overview

  • In high-dimensional statistical inference in which the number of parameters to be estimated is larger than that of the holding data, regularized linear estimation techniques are widely used. These techniques have, however, some drawbacks.

    1. Estimators are biased in the sense that their absolute values are shrunk toward zero because of the regularization effect.
    2. Their statistical properties are difficult to characterize as they are given as numerical solutions to certain optimization problems.
  • In our paper, we tackle such problems concerning LASSO, which is a widely used method for sparse linear estimation, when the measurement matrix is regarded as a sample from a rotationally invariant ensemble. We develop a new computationally feasible scheme to construct a de-biased estimator with a confidence interval.

  • This repository provides a demonstration code for a construction of de-biased estimator.

requirements

  • Python 3 and following Python modules are required to run demonstration notebook
    • numpy
    • scipy
    • matplotlib
    • sklearn

paper

people

About

demo for "A statistical mechanics approach to de-biasing and uncertainty estimation in LASSO for random measurements"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published