Nelder-Mead simplex¶

This solver is implemented in optunity.solvers.NelderMead. It is available in optunity.make_solver() as ‘nelder-mead’.

This is a heuristic, nonlinear optimization method based on the concept of a simplex, originally introduced by Nelder and Mead [NELDERMEAD]. We implemented the version as described on Wikipedia.

This method requires an initial starting point \(x_0\). It is a good local search method, but will get stuck in bad regions when a poor starting point is specified.

Nelder, John A. and Mead, R. A simplex method for function minimization. Computer Journal 7: 308–313, 1965.

Table Of Contents

This Page

Previous topic

Particle Swarm Optimization

Next topic