Improved sign-based learning algorithm derived by the composite nonlinear Jacobi process
From MaRDI portal
Publication:2488877
DOI10.1016/j.cam.2005.06.034zbMath1089.68084OpenAlexW2160070461MaRDI QIDQ2488877
George D. Magoulas, Michael N. Vrahatis, Aristoklis D. Anastasiadis
Publication date: 16 May 2006
Published in: Journal of Computational and Applied Mathematics (Search for Journal in Brave)
Full work available at URL: https://doi.org/10.1016/j.cam.2005.06.034
Convergence analysisGlobal convergenceSupervised learningPattern classificationFeedforward neural networksNonlinear iterative methodsNonlinear Jacobi
Uses Software
Cites Work
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Optimal solution of nonlinear equations
- Empirical evaluation of the improved rprop learning algorithms
- From linear to nonlinear iterative methods
- Learning in multilayer perceptrons using global optimization strategies.
- Algorithm 666: Chabis: a mathematical software package for locating and evaluating roots of systems of nonlinear equations
- On the alleviation of the problem of local minima in back-propagation
- Convergence Conditions for Ascent Methods. II: Some Corrections