×

subplex

swMATH ID: 4818
Software Authors: Tom Rowan; Bruce Lowekamp; Maxime Marmier
Description: SUBPLEX is a subspace-searching simplex method for the unconstrained optimization of general multivariate functions. Like the Nelder-Mead simplex method it generalizes, the subplex method is well suited for optimizing noisy objective functions. The number of function evaluations required for convergence typically increases only linearly with the problem size, so for most applications the subplex method is much more efficient than the simplex method. It can be used like the Matlab fminsearch algorithm. SUBPLEX was developed by Tom Rowan for his Ph.D. Thesis: Functional Stability Analysis of Numerical Algorithms (University of Texas at Austin). Although SUBPLEX was originally developed as a routine for this analysis, it is a general-purpose algorithm well suited for optimization of high-dimensional noisy functions.
Homepage: http://netlib.org/cgi-bin/netlibfiles.pl?filename=/opt/subplex.tgz
Dependencies: Matlab
Related Software: mctoolbox; NLopt; TTDock; BRENT; KELLEY; SUNDIALS; Neper; NASTRAN; MEGAFLOW; Decision tree for optimization software; Algorithm 696; Algorithm 694; Algorithm 710; BHESS; PRECISE; PNM; INTLAB; Mathematica; C-XSC; Matlab
Cited in: 26 Publications

Citations by Year