Results 11 to 20 of about 17,902,926 (294)
Employing a Monte Carlo algorithm in Newton-type methods for restricted maximum likelihood estimation of genetic parameters. [PDF]
PLoS ONE, 2013 Estimation of variance components by Monte Carlo (MC) expectation maximization (EM) restricted maximum likelihood (REML) is computationally efficient for large data sets and complex linear mixed effects models.
Kaarina Matilainen +4 more
doaj +2 more sources
Projected Newton-type Methods in Machine Learning [PDF]
, 2011 We consider projected Newton-type methods for solving large-scale optimization problems arising in machine learning and related fields. We first introduce an algorithmic framework for projected Newton-type methods by reviewing a canonical projected (quasi-)Newton method.
Schmidt, M., Kim, D., Sra, S.
openaire +4 more sources
Newton type iterative methods with higher order of convergence [PDF]
Journal of Numerical Analysis and Approximation Theory, 2016 Newton type iterative methods are obtained with higher order of convergence and with higher efficiency. The methods have been compared with the similar existing methods of recent times.
Pankaj Jain +2 more
doaj +3 more sources
Inexact Nonconvex Newton-Type Methods [PDF]
INFORMS Journal on Optimization, 2021 The paper aims to extend the theory and application of nonconvex Newton-type methods, namely trust region and cubic regularization, to the settings in which, in addition to the solution of subproblems, the gradient and the Hessian of the objective function are approximated. Using certain conditions on such approximations, the paper establishes optimal
Zhewei Yao +3 more
openaire +4 more sources
Gauss–Newton-type methods for bilevel optimization [PDF]
Computational Optimization and Applications, 2021 AbstractThis article studies Gauss–Newton-type methods for over-determined systems to find solutions to bilevel programming problems. To proceed, we use the lower-level value function reformulation of bilevel programs and consider necessary optimality conditions under appropriate assumptions.
Jörg Fliege +2 more
openaire +4 more sources
Newton-type multilevel optimization method [PDF]
Optimization Methods and Software, 2019 Inspired by multigrid methods for linear systems of equations, multilevel optimization methods have been proposed to solve structured optimization problems. Multilevel methods make more assumptions regarding the structure of the optimization model, and as a result, they outperform single-level methods, especially for large-scale models.
Chin Pang Ho +2 more
openaire +4 more sources
A Newton‐type method and its application [PDF]
International Journal of Mathematics and Mathematical Sciences, 2006 We prove an existence and uniqueness theorem for solving the operator equation F(x) + G(x) = 0, where F is a continuous and Gâteaux differentiable operator and the operator G satisfies Lipschitz condition on an open convex subset of a Banach space. As corollaries, a recent theorem of Argyros (2003) and the classical convergence theorem for modified ...
V. Antony Vijesh, P. V. Subrahmanyam
openaire +3 more sources
Hybrid Newton-type method for a class of semismooth equations [PDF]
, 2002 In this paper, we present a hybrid method for the solution of a class of composite semismooth equations encountered frequently in applications. The method is obtained by combining a generalized finite-difference Newton method to an inexpensive direct ...
Pieraccini, Sandra
core +1 more source
Journal of Complexity, 2010 Inexact Newton-type methods are discussed for approximating a locally unique solution of the nonlinear equation \(A(x)^{\#}(F(x)+G(x))=0\) in Banach space. Here \(F\) is a Fréchet-differentiable operator, \(G\) is a continuous operator and \(A(x)^{\#}\) is an analog of the Moore-Penrose generalized inverse of \(A(x)\) which is an approximation of the ...
Ioannis K. Argyros, Saïd Hilout
openaire +3 more sources
FedDANE: A Federated Newton-Type Method [PDF]
2019 53rd Asilomar Conference on Signals, Systems, and Computers, 2019 Asilomar Conference on Signals, Systems, and Computers ...
Tian Li 0005 +5 more
openaire +3 more sources