English
Related papers

Related papers: Quasi-Newton Methods for Machine Learning: Forget …

200 papers

We consider large scale empirical risk minimization (ERM) problems, where both the problem dimension and variable size is large. In these cases, most second order methods are infeasible due to the high cost in both computing the Hessian…

Optimization and Control · Mathematics 2017-05-24 Mark Eisen , Aryan Mokhtari , Alejandro Ribeiro

L-BFGS is the state-of-the-art optimization method for many large scale inverse problems. It has a small memory footprint and achieves superlinear convergence. The method approximates Hessian based on an initial approximation and an update…

Numerical Analysis · Mathematics 2021-03-19 Hari Om Aggrawal , Jan Modersitzki

We consider the development of practical stochastic quasi-Newton, and in particular Kronecker-factored block-diagonal BFGS and L-BFGS methods, for training deep neural networks (DNNs). In DNN training, the number of variables and components…

Machine Learning · Computer Science 2021-01-11 Donald Goldfarb , Yi Ren , Achraf Bahamou

Solving complex optimization problems in engineering and the physical sciences requires repetitive computation of multi-dimensional function derivatives. Commonly, this requires computationally-demanding numerical differentiation such as…

Numerical Analysis · Mathematics 2021-05-12 Danny Smyl , Tyler N. Tallman , Dong Liu , Andreas Hauptmann

We introduce iR2N, a modified proximal quasi-Newton method for minimizing the sum of a smooth function $f$ and a lower semi-continuous prox-bounded function $h$, allowing inexact evaluations of $f$, its gradient, and the associated proximal…

Optimization and Control · Mathematics 2025-12-17 Nathan Allaire , Sébastien Le Digabel , Dominique Orban

Though quasi-Newton methods have been extensively studied in the literature, they either suffer from local convergence or use a series of line searches for global convergence which is not acceptable in the distributed setting. In this work,…

Optimization and Control · Mathematics 2023-12-01 Yubo Du , Keyou You

Newton's method is the most widespread high-order method, demanding the gradient and the Hessian of the objective function. However, one of the main disadvantages of Newtons method is its lack of global convergence and high iteration cost.…

Training deep neural networks for solving machine learning problems is one great challenge in the field, mainly due to its associated optimisation problem being highly non-convex. Recent developments have suggested that many training…

Machine Learning · Computer Science 2017-11-23 Hao Shen

We present a derivative-based algorithm for nonlinearly constrained optimization problems that is tolerant of inaccuracies in the data. The algorithm solves a semi-smooth set of nonlinear equations that are equivalent to the first-order…

Optimization and Control · Mathematics 2017-09-21 Jason E. Hicken , Pengfei Meng , Alp Dener

Models of physics beyond the Standard Model often contain a large number of parameters. These form a high-dimensional space that is computationally intractable to fully explore. Experimental constraints project onto a subspace of viable…

High Energy Physics - Theory · Physics 2022-01-05 Jacob Hollingsworth , Michael Ratz , Philip Tanedo , Daniel Whiteson

Data privacy and security becomes a major concern in building machine learning models from different data providers. Federated learning shows promise by leaving data at providers locally and exchanging encrypted information. This paper…

Machine Learning · Computer Science 2019-12-05 Kai Yang , Tao Fan , Tianjian Chen , Yuanming Shi , Qiang Yang

We consider learning methods based on the regularization of a convex empirical risk by a squared Hilbertian norm, a setting that includes linear predictors and non-linear predictors through positive-definite kernels. In order to go beyond…

Machine Learning · Computer Science 2019-06-19 Ulysse Marteau-Ferey , Dmitrii Ostrovskii , Francis Bach , Alessandro Rudi

This paper proposes a framework of L-BFGS based on the (approximate) second-order information with stochastic batches, as a novel approach to the finite-sum minimization problems. Different from the classical L-BFGS where stochastic batches…

Machine Learning · Computer Science 2018-07-17 Jie Liu , Yu Rong , Martin Takac , Junzhou Huang

We study non-convex empirical risk minimization for learning halfspaces and neural networks. For loss functions that are $L$-Lipschitz continuous, we present algorithms to learn halfspaces and multi-layer neural networks that achieve…

Machine Learning · Computer Science 2015-11-26 Yuchen Zhang , Jason D. Lee , Martin J. Wainwright , Michael I. Jordan

We study randomized sketching methods for approximately solving least-squares problem with a general convex constraint. The quality of a least-squares approximation can be assessed in different ways: either in terms of the value of the…

Optimization and Control · Mathematics 2014-11-04 Mert Pilanci , Martin J. Wainwright

This paper considers consensus optimization problems where each node of a network has access to a different summand of an aggregate cost function. Nodes try to minimize the aggregate cost function, while they exchange information only with…

Optimization and Control · Mathematics 2016-03-24 Mark Eisen , Aryan Mokhtari , Alejandro Ribeiro

Logistic regression is a standard method in multivariate analysis for binary outcome data in epidemiological and clinical studies; however, the resultant odds-ratio estimates fail to provide directly interpretable effect measures. The…

Methodology · Statistics 2024-11-26 Takahiro Kitano , Hisashi Noma

Statistical inverse learning aims at recovering an unknown function $f$ from randomly scattered and possibly noisy point evaluations of another function $g$, connected to $f$ via an ill-posed mathematical model. In this paper we blend…

Statistics Theory · Mathematics 2024-01-22 Tapio Helin

A new result in convex analysis on the calculation of proximity operators in certain scaled norms is derived. We describe efficient implementations of the proximity calculation for a useful class of functions; the implementations exploit…

Optimization and Control · Mathematics 2013-03-04 Stephen Becker , M. Jalal Fadili

This paper proposes a novel class of block quasi-Newton methods for convex optimization which we call symmetric rank-$k$ (SR-$k$) methods. Each iteration of SR-$k$ incorporates the curvature information with~$k$ Hessian-vector products…

Optimization and Control · Mathematics 2024-07-25 Chengchang Liu , Cheng Chen , Luo Luo