Convergence of the reweighted ℓ 1 minimization algorithm for ℓ 2-ℓ p minimization
نویسندگان
چکیده
The iteratively reweighted l1 minimization algorithm (IRL1) has been widely used for variable selection, signal reconstruction and image processing. In this paper, we show that any sequence generated by the IRL1 is bounded and any accumulation point is a stationary point of the l2-lp minimization problem with 0 < p < 1. Moreover, the stationary point is a global minimizer and the convergence rate is approximately linear under certain conditions. We derive posteriori error bounds which can be used to construct practical stopping rules for the algorithm.
منابع مشابه
Compressed sensing for phase retrieval.
To date there are several iterative techniques that enjoy moderate success when reconstructing phase information, where only intensity measurements are made. There remains, however, a number of cases in which conventional approaches are unsuccessful. In the last decade, the theory of compressed sensing has emerged and provides a route to solving convex optimisation problems exactly via ℓ(1)-nor...
متن کاملCuriously Fast Convergence of some Stochastic Gradient Descent Algorithms
1 Context Given a finite set of m examples z 1 ,. .. , z m and a strictly convex differen-tiable loss function ℓ(z, θ) defined on a parameter vector θ ∈ R d , we are interested in minimizing the cost function min θ C(θ) = 1 m m i=1 ℓ(z i , θ). One way to perform such a minimization is to use a stochastic gradient algorithm. Starting from some initial value θ[1], iteration t consists in picking ...
متن کاملMaximum Consensus Parameter Estimation by Reweighted \ell _1 ℓ 1 Methods
Robust parameter estimation in computer vision is frequently accomplished by solving the maximum consensus (MaxCon) problem. Widely used randomized methods for MaxCon, however, can only produce random approximate solutions, while global methods are too slow to exercise on realistic problem sizes. Here we analyse MaxCon as iterative reweighted algorithms on the data residuals. We propose a smoot...
متن کاملApproximate Sparse Decomposition Based on Smoothed L0-Norm
In this paper, we propose a method to address the problem of source estimation for Sparse Component Analysis (SCA) in the presence of additive noise. Our method is a generalization of a recently proposed method (SL0), which has the advantage of directly minimizing the ℓ 0-norm instead of ℓ 1-norm, while being very fast. SL0 is based on minimization of the smoothed ℓ 0-norm subject to As = x. In...
متن کاملCombinatorial Proof of an Abel-type Identity
Identity (1) below resulted from our investigation in [21] of chip-firing games on complete graphs K n , for n ≥ 1; see, e.g., [2] for antecedents. The left side expresses the sum of the probabilities of a game experiencing firing sequences of each possible length ℓ = 0, 1,. .. , n. This note gives a combinatorial proof that these probabilities sum to unity. We first manipulate n − 1 n + 1 + n ...
متن کاملذخیره 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 ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- Comp. Opt. and Appl.
دوره 59 شماره
صفحات -
تاریخ انتشار 2014