Results 1 to 10 of about 97,762 (292)
Continual Learning With Quasi-Newton Methods [PDF]
IEEE Access, 2021 Catastrophic forgetting remains a major challenge when neural networks learn tasks sequentially. Elastic Weight Consolidation (EWC) attempts to address this problem by introducing a Bayesian-inspired regularization loss to preserve knowledge of ...
Steven Vander Eeckt, Hugo Van Hamme
doaj +4 more sources
Matrix Transformations and Quasi-Newton Methods [PDF]
International Journal of Mathematics and Mathematical Sciences, 2007 We first recall some properties of infinite tridiagonal matrices considered as matrix transformations in sequence spaces of the forms sξ, sξ∘, sξ(c), or lp(ξ).
Boubakeur Benahmed +2 more
doaj +4 more sources
Asynchronous Parallel Stochastic Quasi-Newton Methods. [PDF]
Parallel Comput, 2021 Although first-order stochastic algorithms, such as stochastic gradient descent, have been the main force to scale up machine learning models, such as deep neural nets, the second-order quasi-Newton methods start to draw attention due to their effectiveness in dealing with ill-conditioned optimization problems.
Tong Q, Liang G, Cai X, Zhu C, Bi J.
europepmc +5 more sources
Machine Learning in Quasi-Newton Methods
AxiomsIn this article, we consider the correction of metric matrices in quasi-Newton methods (QNM) from the perspective of machine learning theory. Based on training information for estimating the matrix of the second derivatives of a function, we formulate a ...
Vladimir Krutikov +4 more
doaj +2 more sources
Partial Davidon, Fletcher and Powell (DFP) of quasi newton method for unconstrained optimization
Tikrit Journal of Pure Science, 2023 The nonlinear Quasi-newton methods is widely used in unconstrained optimization. However, In this paper, we developing new quasi-Newton method for solving unconstrained optimization problems.
Basheer M. Salih +2 more
doaj +1 more source
Partial Pearson-two (PP2) of quasi newton method for unconstrained optimization
Tikrit Journal of Pure Science, 2023 In this paper, we developing new quasi-Newton method for solving unconstrained optimization problems .The nonlinear Quasi-newton methods is widely used in unconstrained optimization[1]. However,.
Basheer M. Salih +2 more
doaj +1 more source
Mechanical Engineering Journal, 2021 Quasi-Newton-based nonlinear finite element methods were extensively studied in the 1970s and 1980s. However, they have almost disappeared due to their poorer convergence performance than the Newton-Raphson method.
Yasunori YUSA +2 more
doaj +1 more source
A Combined Conjugate Gradient Quasi-Newton Method with Modification BFGS Formula
International Journal of Analysis and Applications, 2023 The conjugate gradient and Quasi-Newton methods have advantages and drawbacks, as although quasi-Newton algorithm has more rapid convergence than conjugate gradient, they require more storage compared to conjugate gradient algorithms.
Mardeen Sh. Taher, Salah G. Shareef
doaj +1 more source
Correlation and realization of quasi-Newton methods of absolute optimization [PDF]
Компьютерные исследования и моделирование, 2016 Newton and quasi-Newton methods of absolute optimization based on Cholesky factorization with adaptive step and finite difference approximation of the first and the second derivatives.
Anastasiya Borisovna Sviridenko +1 more
doaj +1 more source
Decentralized Quasi-Newton Methods [PDF]
IEEE Transactions on Signal Processing, 2017 We introduce the decentralized Broyden-Fletcher-Goldfarb-Shanno (D-BFGS) method as a variation of the BFGS quasi-Newton method for solving decentralized optimization problems. The D-BFGS method is of interest in problems that are not well conditioned, making first order decentralized methods ineffective, and in which second order information is not ...
Mark Eisen +2 more
openaire +2 more sources