Projects

Evolving Objects

EO is a template-based, ANSI-C++ evolutionary computation library which helps you to write your own stochastic optimization algorithms insanely fast. With the help of EO, you can easily design evolutionary algorithms that will find solutions to virtually all kind of hard optimization problems ... [More] , from continuous to combinatorial ones. Designing an algorithm with EO consists in choosing what components you want to use for your specific needs, just as building a structure with Lego blocks. [Less]

evolvotron

This is an interactive generative art application to evolve images/textures/patterns/animations through an iterative process of random mutation and user-selection driven evolution. This process is also often referred to as "evolutionary art" or "genetic art". If you like lava ... [More] lamps, and still think the Mandelbrot set is cool, this could be the software for you. [Less]

MOEA Framework

The MOEA Framework is an open source Java library for developing and experimenting with multiobjective evolutionary algorithms (MOEAs) and other general-purpose optimization algorithms and metaheuristics. A number of algorithms are provided out-of-the-box, including NSGA-II, ε-MOEA, GDE3 and ... [More] MOEA/D. In addition, third-party tools like JMetal and PISA directly integrate with the MOEA Framework. The MOEA Framework targets an academic audience, providing the resources necessary to rapidly design, develop, execute and statistically test optimization algorithms. This includes over 40 test problems from the literature, and a suite of statistical tools for comparing and analyzing algorithm performance. [Less]

GAUL

The Genetic Algorithm Utility Library (or, GAUL for short) is a flexible programming library designed to aid in the development of applications that use genetic, or evolutionary, algorithms. It provides data structures and functions for handling and manipulation of the data required for serial and ... [More] parallel evolutionary algorithms. Additional stochastic algorithms are provided for comparison to the genetic algorithms. [Less]

Java Genetic Algorithm Library

An object oriented library of an Genetic Algorithm, implemented in Java. Clear separation of the several concepts of the algorithm, e.g. Gene, Chromosome, Genotype, Phenotype, Population and Fitness Function. The fitness calculation is parallelized.

Divide and Evolve

A stochastic method for planning decomposition, called Divide-and-Evolve, that was introduced recently and which focuses on plan quality. The basic principle is to search the space of state decompositions of the planning problem at hand by means of artificial evolution: candidate solutions are ... [More] sequences of intermediate goals which define consecutive planning subproblems that are hopefully easier to solve than the global problem. The scope of Divide-and-Evolve is temporal planning as defined by PDDL2.1 (the widely adopted planning domain description language standard) where the problems are described using durative actions and where plan quality is the total makespan. [Less]

EvoSynth

EvoSynth (Evolutionary Computation Synthesizer) is a framework for rapid development and prototyping of evolutionary algorithms.

priska

A collection of Algorithms from my Thesis work. This code is mostly not production-ready or documented and under heavy development. Please contact me if you have specific questions. Thanks, Gabe. Cantor Encoder - A variant of cantor coding. A backpropagation algorithm that allows for the use of ... [More] multiplicative units. Mori, an evolutionary reinforcement algorithm to optimize real-valued vector representable entities. With specific focus on highly partitioned to pseudo-fractal error landscapes in recurrent neural networks. (This project is named after priska, in honor of her telling me forcefully to set up a version-management system.) [Less]

genetik

GenetiK (Genetic Kernel)GenetiK is a generic framework that supports evolutionary algorithms. According to the evolutionary approach, solutions to particular problems can be represented as individuals of a population. For each individual, a particular function, called ''fitness'' ... [More] , can be defined to associate a high fitness scores with 'good' solutions. Starting from any set of individuals, evolving this population (with a process that loosely resembles darwinian selection in biology) results, after a sufficient number of ''generations'', in a new population of high-fitness individuals. In other words, starting from any set (even randomly generated) of potential solutions to a problem, it is possible to obtain new sets of progressively 'better' solutions, eventually resulting in the selection of an optimal (or near-optimal) solution. Therefore, depending on the nature of the problem to be solved, it is possible to find a good solution by expressing candidate solutions and their fitness in an appropriate way. Even adopting the same evolutionary algorithm, but two different representations of individuals and fitness, can lead to dramatically different results, therefore finding a good representation of individuals and their fitness is crucial to the success or failure of a particular application. The idea behind the GenetiK project is to let users experiment with evolutionary computation without having to code from scratch the algorithm that deals with the evolution itself, but rather let them concentrate on their specific tasks and objectives. This allows users to focus on finding good representations for candidate solution, the best ways to measure fitness and other aspects, but, at the same time, allowing them to fine-tune the algorithm to their specific needs by customizing its behaviour, if required. Project StructureThe GenetiK framework is divided in 5 subprojects: genetiK -- that includes the common base classes genetiK::ga -- specialized in dealing with Genetic Algorithms genetiK::gp -- oriented towards Genetic Programming genetiK::gp::st -- dedicated to Strongly Typed Genetic Programming genetiK::util -- includes all the utility classes Here you can find an UML diagram describing the structure of the framework. (Source XMI file, generated with Umbrello) genetiK provides the common evolutionary algorithms functionalities, while genetiK::ga, genetiK::gp and genetiK::gp::st contains the specific classes that should be extended to develop a specific solution. You can consult the annotated example (the genetic algorithm that comes with the library) to see how the framework can be used. DocumentationPlease refer to HTML documentation for more information. Getting the codeTo check out the current version of GenetiK use svn checkout http://genetik.googlecode.com/svn/trunk/ genetikand compile with make(We hope that your current platform is supported ) Running the exampleAfter compiling the library, you can run the example by typing ./bin/example [Less]

Algorithm::Evolutionary

Algorithm::Evolutionary provides classes for performing simple evolutionary computation tasks, including definition of objects from XML and SOAP support. It should be interoperable with other EC libraries using SOAP.