Research Article
Evolutionary and embryogenic approaches to autonomic systems
@INPROCEEDINGS{10.4108/ICST.VALUETOOLS2008.4514, author={Daniele Miorandi and Lidia Yamamoto}, title={Evolutionary and embryogenic approaches to autonomic systems}, proceedings={3rd International ICST Workshop on Interdisciplinary Systems Approach in Performance Evaluation and Design of Computer \& Communication Systems}, publisher={ACM}, proceedings_a={INTER-PERF}, year={2010}, month={5}, keywords={Evolutionary Computation Genetic Programming Genetic Algorithm Artificial Embryogenies Chemical Computing}, doi={10.4108/ICST.VALUETOOLS2008.4514} }- Daniele Miorandi
Lidia Yamamoto
Year: 2010
Evolutionary and embryogenic approaches to autonomic systems
INTER-PERF
ICST
DOI: 10.4108/ICST.VALUETOOLS2008.4514
Abstract
In this paper we present a review of state-of-the-art techniques for automated creation and evolution of software. The focus is on bio-inspired bottom-up approaches, in which complexity is grown from interactions among simpler units. First, we review Evolutionary Computation (EC) techniques, highlighting their potential application to the automated optimization of computer programs in an online, dynamic environment. Then, we survey approaches inspired by embryology, in which artificial entities undergo a developmental process. We introduce the concept of EmbryoWare to refer to software that can be modified via an embryogenic process. We refer to Evolutionary Developmental Computation as the combined evo-devo approach in software, and describe its constituent elements. The paper concludes with a short discussion and outlook for applications of the aforementioned techniques to autonomic computing and communication systems.
