A Java™ based framework for Evolutionary Algorithms -- formerly known as JavaEvA/EvA


EvA2 published on Maven Central Becker - 2015-12-20

The latest version of EvA2 was just published to Maven central. You can now include eva2 as a dependency in your pom.xml using the following XML:


Version Update 2.2.0 Becker - 2015-12-19

We just released version 2.2.0 with some major updates to the graphical user interface, a new command line interface and new optimization methods. The runtime performance was greatly improved. We also updated the documentation to reflect the recent changes.
Please find the new download and updated documentation in the download section. The lastest source code is freely available on our Gitlab instance at https://gitlab.cs.uni-tuebingen.de/eva2/eva2.

Version Update 2.2.0-rc1 Becker - 2014-11-14

Release candidate 1 for EvA2 2.2.0 has been released. It contains many bugfixes to the graphical user interface, a new command line mode and updates to the statistical methods available in EvA2.

Version Update 2.1.0 Becker - 2013-09-01

EvA 2 has undergone a major rewrite in order to clean up the code base. We redesigned the graphical user interface to improve usability and added a new help section to the main menu. New screenshots will be added to the page in the next days.

Thanks to Alexander Seitz we now have added an implementation of the Bayesian Optimization Algorithm to EvA 2.

Version Update 2.05 Kronfeld - 2011-05-05

We have now added Scatter Search, Binary Scatter Search, the Bayesian Optimization Algorithm, and several niching methods to the base package. Among the niching methods, the CBN-PSO method is especially valuable in realistic niching applications. Furthermore, a job list view has been added (thanks to Alex Seitz) and basic statistic tests are supported. To use all of them, the JSC package is required on the classpath.

Version Update 2.046 Kronfeld - 2011-02-02

The new version 2.046 includes extensions on the Matlab interface with explicit data types and support for integer problems next to binary and real-valued. An additional layer of abstraction of individual initialization methods has been introduced. The EvA2 Short Doc has also been updated and the source code examples are delivered in the EvA2BaseSrc package and can directly be run.

Version Update 2.045 Kronfeld - 2010-08-16

The version has been updated to 2.045, with some restructuring, new Terminators and a multi-objective CMA-ES implementation.

Citable Reference Kronfeld - 2010-08-11

The latest citable reference for EvA2 is now complete, presented at the 2010 LION-SWOP (Learning and Intelligent Optimization Conference, Special Session on Software for Optimization), see details (BibTeX, pdf).

Update of SBTOOLBOX2 version Kronfeld - 2010-07-08

We devised an update of the EvA2 interface with SBTOOLBOX2 to work with a more current release. See the download-section.

Adaptions for OpenJDK Kronfeld - 2010-03-02

The latest release should now work without problems with OpenJDK.

Updated Version 2.04 Kronfeld - 2009-11-20

After about half a year we updated the official release of EvA2 to version 2.04. Next to several clean-ups, revisions of gradient descent and the clustering-based niching method, basic command-line arguments are now recognised allowing users to start EvA2 without GUI and with earlier configured run definitions.

Mailing List Installed: Kronfeld - 2008-08-07

We installed a mailing list for users and developers of the EvA2 software package. If you have questions or want to make suggestions on further developement, be welcome to join the EvA mailing list. To subscribe, visit the mailman site at listserv.uni-tuebingen.de.

New Release: Kronfeld - 2008-06-01

The new release version of EvA2 is now online with binary and source packages. The license of EvA2 is LGPL. The project has been split into three branches, making the main optimization kernel, EvA2Base, pretty slim.
New features include a revised GUI, a whole new MATLAB™ interface, several additional optimization benchmarks, and new optimization strategies such as the parameter-free TRIBES implementation (an adaptive swarm) by M.Clerc. Result output is now easier to configure and post-processing can be run easily through the GUI.

Short Introduction: Kronfeld - 2008-05-01

Together with the EvA2 release there is a new Short Introduction to basic functionality and interfaces of EvA2. It is available in the documentation section.

New Name: Kronfeld - 2008-04-01

JavaEvA will, for future versions, be called EvA2 stressing the progress made in functionality and layout during the last time.

JavaEvA Documentation: Streichert - 2005-05-04

The JavaEvA documentation has been published as technical report, can now be cited and is online available.

JavaEvA won 2nd price of the doIT-Award: Streichert - 2004-11-02

doIT Software-Award & doIT Software-Forschungstag

Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners.

The brand name MATLAB is a registered trademark of The Mathworks, Inc. in the United States and other countries.