Section Calculation Modes::Geometry Optimization
Method by which the minimization is performed. For more information see the GSL documentation.
Simple steepest descent.
(Experimental) Non-gsl implementation of steepest descent.
Fletcher-Reeves conjugate-gradient algorithm. The
conjugate-gradient algorithm proceeds as a succession of line
minimizations. The sequence of search directions is used to build
up an approximation to the curvature of the function in the
neighborhood of the minimum.
Polak-Ribiere conjugate-gradient algorithm.
Vector Broyden-Fletcher-Goldfarb-Shanno (BFGS) conjugate-gradient algorithm.
It is a quasi-Newton method which builds up an approximation to the second
derivatives of the function f using the difference between successive gradient
vectors. By combining the first and second derivatives, the algorithm is able
to take Newton-type steps towards the function minimum, assuming quadratic
behavior in that region.
The bfgs2 version of this minimizer is the most efficient version available,
and is a faithful implementation of the line minimization scheme described in
Fletcher, Practical Methods of Optimization, Algorithms 2.6.2 and 2.6.4.
This is experimental, and in fact, not recommended unless you just want to
fool around. It is the Nead-Melder simplex algorithm, as implemented in the
GNU Scientific Library (GSL). It does not make use of the gradients (i.e., the
forces) which makes it less efficient than other schemes. It is included here
for completeness, since it is free.
The FIRE algorithm. See also GOFireMass and GOFireIntegrator.
Ref: E. Bitzek, P. Koskinen, F. Gahler, M. Moseler, and P. Gumbsch, Phys. Rev. Lett. 97, 170201 (2006).
main/geom_opt.F90 : 331
call parse_variable(sys%namespace, 'GOMethod', MINMETHOD_FIRE, g_opt%method)
Featured in tutorials
Featured in chapters of the manual:
Featured in testfiles