Exact Recovery of Hard Thresholding Pursuit

نویسندگان

  • Xiao-Tong Yuan
  • Ping Li
  • Tong Zhang
چکیده

The Hard Thresholding Pursuit (HTP) is a class of truncated gradient descent methods for finding sparse solutions of l0-constrained loss minimization problems. The HTP-style methods have been shown to have strong approximation guarantee and impressive numerical performance in high dimensional statistical learning applications. However, the current theoretical treatment of these methods has traditionally been restricted to the analysis of parameter estimation consistency. It remains an open problem to analyze the support recovery performance (a.k.a., sparsistency) of this type of methods for recovering the global minimizer of the original NP-hard problem. In this paper, we bridge this gap by showing, for the first time, that exact recovery of the global sparse minimizer is possible for HTP-style methods under restricted strong condition number bounding conditions. We further show that HTP-style methods are able to recover the support of certain relaxed sparse solutions without assuming bounded restricted strong condition number. Numerical results on simulated data confirms our theoretical predictions.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Hard Thresholding Pursuit Algorithms: Number of Iterations

The Hard Thresholding Pursuit algorithm for sparse recovery is revisited using a new theoretical analysis. The main result states that all sparse vectors can be exactly recovered from compressive linear measurements in a number of iterations at most proportional to the sparsity level as soon as the measurement matrix obeys a certain restricted isometry condition. The recovery is also robust to ...

متن کامل

Sparse Recovery Algorithms: Sufficient Conditions in terms of Restricted Isometry Constants

We review three recovery algorithms used in Compressive Sensing for the reconstruction s-sparse vectors x ∈ CN from the mere knowledge of linear measurements y = Ax ∈ Cm, m < N. For each of the algorithms, we derive improved conditions on the restricted isometry constants of the measurement matrix A that guarantee the success of the reconstruction. These conditions are δ2s < 0.4652 for basis pu...

متن کامل

A Study on Sparse Vector Distributions and Recovery from Compressed Sensing

Bob L. Sturm, Member, IEEE, Abstract I empirically investigate the variability of several recovery algorithms on the distribution underlying the sparse vector sensed by a random matrix. a dependence that has been noted before, but, to my knowledge, not thoroughly investigated. I find that `1-minimization [1] and tuned two-stage thresholding [2] (subspace pursuit [3] without the use of a sparsit...

متن کامل

Comparison of threshold-based algorithms for sparse signal recovery

Intensively growing approach in signal processing and acquisition, the Compressive Sensing approach, allows sparse signals to be recovered from small number of randomly acquired signal coefficients. This paper analyses some of the commonly used threshold-based algorithms for sparse signal reconstruction. Signals satisfy the conditions required by the Compressive Sensing theory. The Orthogonal M...

متن کامل

Iterative methods for random sampling and compressed sensing recovery

In this paper, two methods are proposed which address the random sampling and compressed sensing recovery problems. The proposed random sampling recovery method is the Iterative Method with Adaptive Thresholding and Interpolation (IMATI). Simulation results indicate that the proposed method outperforms existing random sampling recovery methods such as Iterative Method with Adaptive Thresholding...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2016