Imitation programming

  • Authors:
  • Larry Bull

  • Affiliations:
  • Department of Computer Science, University of the West of England, Bristol, UK

  • Venue:
  • ICES'10 Proceedings of the 9th international conference on Evolvable systems: from biology to hardware
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Many nature-inspired mechanisms have been presented for computational design and optimization. This paper introduces a population-based approach inspired by a form of cultural learning - imitation. Imitation is typically defined as learning through the copying of others. In particular, it is used in this paper to design simple circuits using a discrete dynamical system representation - Turing's unorganised machines. Initial results suggest the imitation computation approach presented is competitive with evolutionary computation, i.e., another class of stochastic population-based search, to design circuits from such recurrent NAND gate networks. Synchronous and asynchronous circuits are considered.