In this report, we propose a primal interior-point method for large sparse generalized minimax optimization. After a short introduction, where the problem is stated, we introduce the basic equations of the Newton method applied to the KKT conditions and propose a primal interior-point method. Next we describe the basic algorithm and give more details concerning … Read more
In this paper, we study convex optimization methods for computing the trace norm regularized least squares estimate in multivariate linear regression. The so-called factor estimation and selection (FES) method, recently proposed by Yuan et al. [17], conducts parameter estimation and factor selection simultaneously and have been shown to enjoy nice properties in both large and … Read more
By introducing some redundant Klee-Minty constructions, we have previously shown that the central path may visit every vertex of the Klee-Minty cube having $2^n-2$ “sharp” turns in dimension $n$. In all of the previous constructions, the maximum of the distances of the redundant constraints to the corresponding facets is an exponential number of the dimension … Read more
In this paper we analyze the limiting behavior of infeasible weighted central paths in semidefinite programming under the assumption that a strictly complementary solution exists. We show that the paths associated with the “square root” symmetrization of the weighted centrality condition are analytic functions of the barrier parameter $\mu$ even at $\mu=0$ if and only … Read more
In this paper, we study polynomial-time interior-point algorithms in view of information geometry. Information geometry is a differential geometric framework which has been successfully applied to statistics, learning theory, signal processing etc. We consider information geometric structure for conic linear programs introduced by self-concordant barrier functions, and develop a precise iteration-complexity estimate of the polynomial-time … Read more
We analyze the properties of an interior point cutting plane algorithm that is based on a semi-infinite linear formulation of the dual semidefinite program. The cutting plane algorithm approximately solves a linear relaxation of the dual semidefinite program in every iteration and relies on a separation oracle that returns linear cutting planes. We show that … Read more
We describe a specialized truncated-Newton primal-dual interior-point method that solves large scale network utility maximization problems, with concave utility functions, efficiently and reliably. Our method is not decentralized, but easily scales to problems with a million flows and links. We compare our method to a standard decentralized algorithm based on dual decomposition, and show by … Read more
A regularized Newton-like method for solving nonnegative least-squares problems is proposed and analysed in this paper. A preconditioner for KKT systems arising in the method is introduced and spectral properties of the preconditioned matrix are analysed. A bound on the condition number of the preconditioned matrix is provided. The bound does not depend on the … Read more
Linear support vector machine training can be represented as a large quadratic program. We present an efficient and numerically stable algorithm for this problem using interior point methods, which requires only O(n) operations per iteration. Through exploiting the separability of the Hessian, we provide a unified approach, from an optimization perspective, to 1-norm classification, 2-norm … Read more
In this work we develop a primal-dual short-step interior point method for conic convex optimization problems for which exact evaluation of the gradient and Hessian of the barrier function is either impossible or too expensive. As our main contribution, we show that if approximate gradients and Hessians can be computed, and the relative errors in … Read more