FIRD - Filter Inexact Restoration Derivative-free algorithm

This algorithm is based on the paper

Ferreira, P. S., Karas, E. W., Sachine, M., & Sobral, F. N. C. (2017). Global convergence of a derivative-free inexact restoration filter algorithm for nonlinear programming. Optimization, 66(2), 271–292.

FIRD is able to solve general derivative-free constrained optimization problems in which the derivatives of the objective function are not available.

How to install

In the provided version, FIRD uses a modified version of the algorithm TRDF for building and updating the trust region models. To solve the nonlinear trust-region problems, it uses ALGENCAN.

Installing ALGENCAN

To install and build ALGENCAN library

1. Download ALGENCAN;

2. Unpack it and, in the main folder, e.g $ALGSRC, simply type make in order to generate $ALGSRC/lib/libalgencan.a file;

Installing and building FIRD

1. Download and unpack FIRD;

2. In the main subdirectory, edit file Makefile with the path to ALGENCAN library

    SOLVERLIB = $ALGSRC/lib
   

   where $ALGSRC is the name of the full path where ALGENCAN has been installed;

3. Type

    make fird
   

   to generate test problem HS14 from the Hock-Schittkowski collection. The executable file will be generated in /bin subdirectory. Or, type

    make fird PROBLEM=tests/examples/toyprob.f90
   

   for a small example.

4. Run it

    ./bin/fird
   

Advanced users

FIRD can be modified in order to use different solvers for

• Feasibility phase
• Construction and updating models (optimality phase)
• Solving nonlinear programming trust-region subproblems (optimality phase)

The manual for advanced information can be generated with the command

make doc