The First International Symposium on
Nature-Inspired Systems for Parallel, Asynchronous and Decentralised Environments
NISPADE
April 4th, 2006
part of
AISB '06: Adaptation in Artificial and Biological Systems
University of Bristol, Bristol, England
Overview
Nature-inspired algorithms such as genetic algorithms, particle swarm optimisation and ant colony algorithms are the state-of-the-art solution technique for some problems. Furthermore, their population-based stochastic search approach promises desirable algorithm features such as anytime decentralised solution and robustness to problem change. However, the efficient pursuit of more accurate solutions leads researchers to appeal to centralised, highly tuned and sequential implementations that are only loosely related to their successful natural counterparts. This renders them brittle in the face of the dynamism of changing problem specifications and operating conditions and limits their usefulness to industry's direction of increasing distribution, decentralisation and adaptability.
Emerging computing environments such as autonomic computing, ubiquitous computing, Peer-to-Peer systems, the Grid and the Semantic Web demand the interaction of large numbers of decentralised, parallel, asynchronous, and distributed software entities in a standardised fashion.
If nature-inspired algorithms are to make an impact on these emerging computing environments, disciplined scientific and engineering investigations must be undertaken into the successful transfer of these algorithms, techniques and infrastructures into such environments.
Topics of Interest
Topics of interest include (but are not limited to):
Methodologies:
Searching the vast parameter spaces of nature-inspired systems.
Empirical performance evaluation and benchmarking procedures.
Design and programming abstractions to manage the complexity of parallel, asynchronous and decentralised nature-inspired techniques
Software engineering techniques for parallel, asynchronous and decentralised
nature-inspired systems, e.g., design patterns, component frameworks and software architectures
Middleware for implementing algorithms in MASs:
Supporting nature-inspired algorithms in a decentralised, asynchronous and parallel context (e.g. pheromone infrastructures).
Integrating implementations within existing middleware technologies.
Ontologies and protocols for nature-inspired system functionality (e.g. pheromone deposition, aggregation and dispersion).
Applications:
Design and development of parallel, asynchronous and decentralised applications using nature-inspired techniques
Applications of nature-inspired techniques in novel areas, such as (but not
limited to) mobile, pervasive and grid computing
Scalability and performance optimisation of parallel, asynchronous and decentralised nature-inspired
applications
Tool support for nature-inspired techniques
Experiences And Results:
New issues in the emerging computing environments context (e.g. asynchronicity, self-organisation, hyperactivity, agent redundancy, messaging costs).
Measures of the above.
Efficiency, robustness, population diversity, adaptiveness and other qualities of nature-inspired implementations.
Submissions
Papers should be 4 pages in length, PDF format, formatted according to the AISB
formatting guidelines available under the 'Authors' section of the symposium web
site www.aisb.org.uk/convention/aisb06/authors.html
Please submit your paper by e-mail to ERidge@cs.york.ac.uk
with the subject line clearly
identifying "NISPADE 2006 Submission". Your submission e-mail must
contain the PDF file as a MIME attachment. Only PDF format files can be
accepted. The sender of the submission will be
the contact person, unless otherwise requested in the submission.
Publication
Papers will be accepted as 'Long' papers and 'Short' papers for the final
symposium proceedings which will be published by the AISB. Short papers will be accepted at the
4 page submission length. Long
papers may be up to 8 pages.
Selected researchers taking part in NISPADE
2006 will be invited to submit a significantly extended version
of their symposium contribution for consideration for a special issue of the international journal Multiagent
and Grid Systems.
