An Investigation of Newton-Sketch and Subsampled Newton Methods

Sketching, a dimensionality reduction technique, has received much attention in the statistics community. In this paper, we study sketching in the context of Newton’s method for solving finite-sum optimization problems in which the number of variables and data points are both large. We study two forms of sketching that perform dimensionality reduction in data space: … Read more

Exact and Inexact Subsampled Newton Methods for Optimization

The paper studies the solution of stochastic optimization problems in which approximations to the gradient and Hessian are obtained through subsampling. We first consider Newton-like methods that employ these approximations and discuss how to coordinate the accuracy in the gradient and Hessian to yield a superlinear rate of convergence in expectation. The second part of … Read more

Optimization Methods for Large-Scale Machine Learning

This paper provides a review and commentary on the past, present, and future of numerical optimization algorithms in the context of machine learning applications. Through case studies on text classification and the training of deep neural networks, we discuss how optimization problems arise in machine learning and what makes them challenging. A major theme of … Read more

A Stochastic Quasi-Newton Method for Large-Scale Optimization

Abstract The question of how to incorporate curvature information in stochastic approximation methods is challenging. The direct application of classical quasi- Newton updating techniques for deterministic optimization leads to noisy curvature estimates that have harmful effects on the robustness of the iteration. In this paper, we propose a stochastic quasi-Newton method that is efficient, robust … Read more

An Inexact Successive Quadratic Approximation Method for Convex L-1 Regularized Optimization

We study a Newton-like method for the minimization of an objective function $\phi$ that is the sum of a smooth convex function and an $\ell_1$ regularization term. This method, which is sometimes referred to in the literature as a proximal Newton method, computes a step by minimizing a piecewise quadratic model $q_k$ of the objective … Read more

A Family of Second-Order Methods for Convex L1-Regularized Optimization

This paper is concerned with the minimization of an objective that is the sum of a convex function $f$ and an $\ell_1$ regularization term. Our interest is in methods that incorporate second-order information about the function $f$ to accelerate convergence. We describe a semi-smooth Newton framework that can be used to generate a variety of … Read more

Newton-Like Methods for Sparse Inverse Covariance Estimation

We propose two classes of second-order optimization methods for solving the sparse inverse covariance estimation problem. The first approach, which we call the Newton-LASSO method, minimizes a piecewise quadratic model of the objective function at every iteration to generate a step. We employ the fast iterative shrinkage thresholding method (FISTA) to solve this subproblem. The … Read more

Sample Size Selection in Optimization Methods for Machine Learning

This paper presents a methodology for using varying sample sizes in batch-type optimization methods for large scale machine learning problems. The first part of the paper deals with the delicate issue of dynamic sample selection in the evaluation of the function and gradient. We propose a criterion for increasing the sample size based on variance … Read more

Subspace accelerated matrix splitting algorithms for bound-constrained quadratic programming and linear complementarity problems

This paper studies the solution of two problems—bound-constrained quadratic programs and linear complementarity problems—by two-phase methods that consist of an active set prediction phase and a subspace phase. The algorithms enjoy favorable convergence properties under weaker assumptions than those assumed for other methods in the literature. The active set prediction phase employs matrix splitting iterations … Read more

An Implementation of an Algorithm for Nonlinear Programming Based on Piecewise Linear Models

This is a progress report on an implementation of the active-set method for nonlinear programming proposed in [6] that employs piecewise linear models in the active-set prediction phase. The motivation for this work is to develop an algorithm that is capable of solving large-scale problems, including those with a large reduced space. Unlike SQP methods, … Read more