Optimization Algorithm Toolkit Publisher's description
from Jason Brownlee
Optimization Algorithm Toolkit is a workbench and toolkit for developing, evaluating, and playing with optimization algorithms.
Optimization Algorithm Toolkit is a workbench and toolkit for developing, evaluating, and playing with classical and state-of-the-art optimization algorithms on standard benchmark problem domains; including reference algorithm implementations, graphing, visualizations and much more. The project was developed initially by Jason Brownlee as a part of his Ph.D. program.
The goal of this project is to deliver obscure-and-prevalent, old-and-new optimization algorithms from research literature to both research scientists and algorithm practitioners alike. Algorithms include biologically inspired approaches such as evolutionary algorithms (genetic algorithms), swarm algorithms (ants and particle swarm), and immune system algorithms.
Also included are more conventional approaches such as approaches inspired by physics including simulated annealing and extremal optimization. Problem domains include numerical function optimization, traveling salesman problems, and protein folding all with many standard benchmark instances taken from research literature.
A user-friendly graphical interface is provided to rapidly evaluate and compare algorithm and problem configurations, visualize algorithm behavior, and graph algorithm performance over time. A robust, modular, and extensible framework underlies the platform to facilitate the easy addition and modification of algorithms, addition of new problem domains and problem instances as well as facilitate more advanced algorithm experimentation.
The algorithm implementations are extensible and easily support modification and applicaition to varied problem domains. Please report any bugs, feature requests or include your own algorithms by accessing the services on the project home website. This is an open source project (released under the GPL) so the source code is available. The project was compiled with Java 1.5 (update 9).
What's New in This Release:?пїЅ A major restructuring of the API was done.
?пїЅ Bugs were fixed everywhere.
?пїЅ A new (and beta) experimenter API and graphical user interface were added.
?пїЅ Notably, the new experimenter includes the standard statistical hypothesis tests for normality and comparison of algorithm results.
System Requirements:No special requirements.
Program Release Status: Minor Update
Program Install Support: Install and Uninstall