Evolving Bio-PEPA process algebra models using genetic programming

  • Authors:
  • David Marco;Carron Shankland;David Cairns

  • Affiliations:
  • University of Stirling, Stirling, United Kingdom;University of Stirling, Stirling, United Kingdom;University of Stirling, Stirling, United Kingdom

  • Venue:
  • Proceedings of the 14th annual conference on Genetic and evolutionary computation
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper presents initial results of applying a Genetic Programming (GP) approach to the evolution of process algebra models defined in Bio-PEPA. An incomplete model of a system is provided together with target behaviour. GP is then used to evolve new definitions that complete the model while ensuring a good fit to target data. Our results show that a set of effective models can be developed with this approach that can either be used directly or further refined using a modeller's domain knowledge. Such an approach can greatly reduce the time taken to develop new models, enabling a modeller to focus on the subtler modelling aspects of the problem domain. Although the work presented here concerns the modelling of biological systems, the approach is generally applicable to systems for which appropriate target behaviour can be captured and that can be formalised as a set of communicating processes.