The NEOS Server offers ACRS (Adaptive Controlled Random Search) algorithm for the solution of global optimization problems of small to medium size (say, a few tens of variables).
For a more complete description of ACRS see:
ACRS has been developed by G. Di Pillo, G. Liuzzi, and S. Lucidi at the Department of Computer and System Sciences (DIS) "A. Ruberti", University of Rome "La Sapienza".
Additional information can be obtained by sending an e-mail message to G. Liuzzi
ACRS is supported by MIUR, FIRB Research Program Large-Scale Nonlinear Optimization, Rome, Italy.
The model is specified by a model file, and optionally, a data file and a commands file. If the command file is specified it must contain the AMPL solve command. The model file and the data file must not contain the solve command.
The command file can contain any AMPL command or set options for ACRS.
ACRS accepts the following options: cpulim - which limits the maximum allowed cpu time (default 3600.0 sec) maxgen - which limits the maximum allowed time for workspace generation (default 300.0 sec) maxit - which limits the maximum number of iterations (default 100,000) outlev - output level, ranging from -1 to 2 for increasing output verbosity (default -1) tol - tolerance in the stopping criterion (default 1.0e-5) wkdimen - working set dimension (must be >= N+1, default max[50,25*N]).
ACRS options can be set by using the usual AMPL syntax: option acrs_options 'option=val';
DOE disclaimer DOE Web privacy policy