A Java-based Evolutionary Computation Research System
By Sean Luke, Eric O. Scott, Liviu Panait, Gabriel Balan, Sean Paus, Zbigniew Skolicki, Rafal Kicinger, Elena Popovici, Keith Sullivan, Joseph Harrison, Jeff Bassett, Robert Hubley, Ankur Desai, Alexander Chircop, Jack Compton, William Haddon, Stephen Donnelly, Beenish Jamil, Joseph Zelibor, Eric Kangas, Faisal Abidi, Houston Mooers, James O'Beirne, Khaled Ahsan Talukder, Sam McKay, James McDermott, Jason Zou, Anson Rutherford, David Freelan, Ermo Wei, Sunil Rajendran, Ananya Dhawan, Ben Brumbac, Anson Rutherford.
ECJ is a research EC system written in Java. It was designed to be highly flexible, with nearly all classes (and all of their settings) dynamically determined at runtime by a user-provided parameter file. All structures in the system are arranged to be easily modifiable. Even so, the system was designed with an eye toward efficiency.
ECJ is developed at George Mason University's ECLab Evolutionary Computation Laboratory. The software has nothing to do with its initials' namesake, Evolutionary Computation Journal. ECJ's sister project is MASON, a multi-agent simulation system which dovetails with ECJ nicely.
The new directions for ECJ (starting with Version 25) were presented at GECCO 2017 in the paper ECJ Then and Now.
The ECJ team has received a three-year NSF grant to improve and enhance ECJ. Version 26 is the second release under this grant, and reflects a lot of additions.
- GUI with charting
- Platform-independent checkpointing and logging
- Hierarchical parameter files
- Mersenne Twister Random Number Generators
- Abstractions for implementing a variety of EC forms.
- A really, really big manual
- Optional builds with Apache Maven
- Asynchronous island models over TCP/IP
- Master/Slave evaluation over multiple processors, with support for generational, asynchronous steady-state, and coevolutionary distribution
- Genetic Algorithms/Programming style Steady State and Generational evolution, with or without Elitism
- Evolutionary-Strategies style (mu,lambda) and (mu+lambda) evolution
- Dedicated package for efficient single-state (hill-climbing, simulated annealing, etc.) methods
- CMA-ES, AMaLGaM IDEA, PBIL, DOvs EDAs
- Very flexible breeding architecture
- Many selection operators
- Multiple subpopulations and species
- Inter-subpopulation exchanges
- Reading populations from files
- Single- and Multi-population coevolution
- NSGA-II, NSGA-III, and SPEA2 multiobjective optimization
- Particle Swarm Optimization
- Differential Evolution
- Spatially embedded evolutionary algorithms
- Hooks for other multiobjective optimization methods
- Packages for parsimony pressure
|GP Tree Representations
Vector (GA/ES) Representations
- Set-based Strongly-Typed Genetic Programming
- Ephemeral Random Constants
- Automatically-Defined Functions and Automatically Defined Macros
- Multiple tree forests
- Six tree-creation algorithms
- Extensive set of GP breeding operators
- Grammatical Encoding
- Many pre-done GP application problem domains, including ant, regression, multiplexer, lawnmower, parity, two-box, edge
- Fixed-Length and Variable-Length Genomes
- Arbitrary representations
- Variety of mutation and crossover operators
- Many pre-done vector application problem domains (rastrigin, sum, rosenbrock, sphere, step, noisy-quartic, booth, griewangk, nk, hiff, median)
- 23 Black-Box Optimization competition benchmark problems, plus noisy versions.
- Multiset-based genomes in the rule package, for evolving Pitt-approach rulesets or other set-based representations.
Download ECJ version 26 as ecj26.tar.gz or (bigger) ecj26.zip.
Download ECJ's support libraries as libraries.tar.gz or libraries.zip. The support libraries provide four facilities:
- ECJ's CMA-ES code relies on the matrix library EJML.
- ECJ's distributed evaluation and island model facilities have optional compressed socket options for more efficiency. Java's standard compression routines are broken for sockets (they don't support "partial flush") and so we rely on the JZlib library to do it.
- ECJ's GUI relies on the JFreeChart and iText libraries, specifically the jfreechart.jar, jcommon.jar, and iText.jar files. Use the ECJ versions, not the versions on those websites. If you don't wish to use the GUI and thus install JFreeChart and iText, you can just get rid of the ec/display directory and don't call make gui, and ECJ will compile and run from the command line fine. Per license agreement, we provide source to these libraries here.
- ECJ's Push facility relies on a modified version of the Psh interpreter. Be sure to use the ECJ version.
New! The ECJ Owner's Manual is now available. Hundreds of pages of mind-numbing tedium detailing every last feature. Enjoy!
The full online documentation of the system, including tutorials, is also available.
Repository: ECJ's repository is at Github, under the project name ecj.
Contrib packages: (also out on CVS in the contrib directory)
- CGP by David Oranchak: contrib-cgp-18.zip. Extensions for Julian Miller's Cartesian Genetic Programming. Documentation and tutorial here.
- DRM by Alberto Cuesta: drm.zip. Extensions to marry run ECJ with the DRM peer-to-peer distributed computation engine developed for the dr-ea-m distributed EC project.
- GEP by Bob Orchard: ecj18Gep.zip. Extensions for Gene Expression Programming.
- Xholon by Ken Webb: Simple modifications to ECJ that allow it to be embedded within the Xholon modeling and simulation tool.
- Teambots ECJ by Liviu Panait and Sean Luke: teambots.tar.gz or teambots.zip. Extensions to ECJ to make it interoperate reasonably well with the venerable TeamBots lightweight robot simulation package. Getting up in age: we don't use it any longer.
Genetic Programming and Evolvable Machines has reviewed ECJ, and kindly, a few times. The most recent is "Software Review: the ECJ Toolkit" by David White.
If you have questions or ideas regarding ECJ, we suggest you join the ECJ-INTEREST Discussion List. (Alternatively, send mail to firstname.lastname@example.org with the words subscribe ECJ-INTEREST-L in the body of the message. Likewise, to unsubscribe, use unsubscribe ECJ-INTEREST-L). You can also view the archives. If you want to report a bug, you can contact the ECJ authors directly at ecj-help @ cs.gmu.edu