Evolution and Acquisition of Modules in Cartesian Genetic Programming

Walker, James Alfred; Miller, Julian F.

doi:10.1007/978-3-540-24650-3_17

Cited by 61 publications

(62 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Further, CGP models have been used with different evolutionary strategies, from classic genetic algorithms, e.g., [4], to evolutionary strategies, e.g., [28,29]. An often-used mutation operator on the CGP model is one-point mutation.…”

Section: Cartesian Genetic Programsmentioning

confidence: 99%

See 1 more Smart Citation

Advanced techniques for the creation and propagation of modules in cartesian genetic programming

Kaufmann

Platzner

2008

Proceedings of the 10th Annual Conference on Genetic and Evolutionary Computation

View full text Add to dashboard Cite

The choice of an appropriate hardware representation model is key to successful evolution of digital circuits. One of the most popular models is cartesian genetic programming, which encodes an array of logic gates into a chromosome. While several smaller circuits have been successfully evolved on this model, it lacks scalability. A recent approach towards scalable hardware evolution is based on the automated creation of modules from primitive gates.In this paper, we present two novel approaches for module creation, an age-based and a cone-based technique. Further, we detail a cone-based crossover operator for use with cartesian genetic programming. We evaluate the different techniques and compare them with related work. The results show that age-based module creation is highly effective, while cone-based approaches are only beneficial for regularly structured, multiple output functions such as multipliers.

show abstract

Section: Cartesian Genetic Programsmentioning

confidence: 99%

“…The main contribution of this paper is the introduction of novel techniques for identifying and dealing with modules, leveraging a previously discussed approach for module creation in the CGP model [28]. We present and evaluate two module creation techniques, called age-based and conebased module creation.…”

Section: Introductionmentioning

confidence: 99%

Advanced techniques for the creation and propagation of modules in cartesian genetic programming

Kaufmann

Platzner

2008

Proceedings of the 10th Annual Conference on Genetic and Evolutionary Computation

View full text Add to dashboard Cite

show abstract

“…An evolved circuit consists of a number of primary inputs, a number of logic blocks, and a number of primary outputs. ECGP extends CGP by relaxing the strict geometric layout constraints and by adding the automatic definition and reuse of sub-functions (modules) [13]. While primitive nodes correspond to basic gate functions, modules are defined as compositions of primitive nodes.…”

Section: The Embedded Cartesian Genetic Programming (Ecgp)mentioning

confidence: 99%

“…As a fitness metric for the training phase we use the reciprocal square error distance to the predictions of a perfect classifier. Similar to [13], we have chosen a standard 1 + 4 evolutionary strategy (ES) as the optimization algorithm. The population is initialized randomly with circuits that comprise 10 logic blocks.…”

Section: The Embedded Cartesian Genetic Programming (Ecgp)mentioning

confidence: 99%

A Comparison of Evolvable Hardware Architectures for Classification Tasks

Glette

Tørresen

Kaufmann

et al. 2008

Evolvable Systems: From Biology to Hardware

View full text Add to dashboard Cite

Abstract. We analyze and compare four different evolvable hardware approaches for classification tasks: An approach based on a programmable logic array architecture, an approach based on two-phase incremental evolution, a generic logic architecture with automatic definition of building blocks, and a specialized coarse-grained architecture with pre-defined building blocks. We base the comparison on a common data set and report on classification accuracy and training effort. The results show that classification accuracy can be increased by using modular, specialized classifier architectures. Furthermore, function level evolution, either with predefined functions derived from domain-specific knowledge or with functions that are automatically defined during evolution, also gives higher accuracy. Incremental and function level evolution reduce the search space and thus shortens the training effort.

show abstract

“…The principle of automatically defined functions (ADFs) from GP [21] has been applied to the evolution of analog circuits [19]. An alternative approach, called embedded cartesian genetic programming (ECGP), has been applied to the evolution of digital circuits [44]. Both approaches select portions from a solution at random and create building blocks from them.…”

Section: Scalabilitymentioning

confidence: 99%