An $\mathcal{O}(n\log n)$ projection operator for weighted $\ell_1$-norm regularization with sum constraint
نویسنده
چکیده
We provide a simple and efficient algorithm for the projection operator for weighted l1-norm regularization subject to a sum constraint, together with an elementary proof. The implementation of the proposed algorithm can be downloaded from the author’s homepage. 1 The problem In this report, we consider the following optimization problem: min x 1 2 ‖x− y‖ 2 + n
منابع مشابه
An O(n log n) projection operator for weighted ℓ1-norm regularization with sum constraint
We provide a simple and efficient algorithm for the projection operator for weighted l1-norm regularization subject to a sum constraint, together with an elementary proof. The implementation of the proposed algorithm can be downloaded from the author’s homepage. 1 The problem In this report, we consider the following optimization problem: min x 1 2 ‖x − y‖ 2 + n
متن کاملAn O(nlog(n)) Algorithm for Projecting Onto the Ordered Weighted ℓ1 Norm Ball
The ordered weighted `1 (OWL) norm is a newly developed generalization of the Octogonal Shrinkage and Clustering Algorithm for Regression (OSCAR) norm. This norm has desirable statistical properties and can be used to perform simultaneous clustering and regression. In this paper, we show how to compute the projection of an n-dimensional vector onto the OWL norm ball in O(n log(n)) operations. I...
متن کاملGeneralized Weighted Composition Operators From Logarithmic Bloch Type Spaces to $ n $'th Weighted Type Spaces
Let $ mathcal{H}(mathbb{D}) $ denote the space of analytic functions on the open unit disc $mathbb{D}$. For a weight $mu$ and a nonnegative integer $n$, the $n$'th weighted type space $ mathcal{W}_mu ^{(n)} $ is the space of all $fin mathcal{H}(mathbb{D}) $ such that $sup_{zin mathbb{D}}mu(z)left|f^{(n)}(z)right|begin{align*}left|f right|_{mathcal{W}_...
متن کاملAn $O(n\log(n))$ Algorithm for Projecting Onto the Ordered Weighted $\ell_1$ Norm Ball
The ordered weighted `1 (OWL) norm is a newly developed generalization of the Octogonal Shrinkage and Clustering Algorithm for Regression (OSCAR) norm. This norm has desirable statistical properties and can be used to perform simultaneous clustering and regression. In this paper, we show how to compute the projection of an n-dimensional vector onto the OWL norm ball in O(n log(n)) operations. I...
متن کاملDecreasing Weighted Sorted $\ell_1$ Regularization
We consider a new family of regularizers, termed weighted sorted `1 norms (WSL1), which generalizes the recently introduced octagonal shrinkage and clustering algorithm for regression (OSCAR) and also contains the `1 and `∞ norms as particular instances. We focus on a special case of the WSL1, the decreasing WSL1 (DWSL1), where the elements of the argument vector are sorted in non-increasing or...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2015