[Please circulate to all those who might be interested, and accept our apologies if you received multiple copies of this announcement] - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Call for Extended Abstracts Workshop on Systems to Build Systems July 18, 2010, Barcelona, SPAIN As a part of 2010 IEEE World Congress on Computational Intelligence http://www.wcci2010.org/ Deadline: June 7, 2010 Instructions for authors: The Organising Committee invites you to submit your extended abstract in PDF format to [log in to unmask] by the deadline June 7, 2010. The extended abstract should not be longer than 2 pages or 1000 words (with references included). The authors of the selected papers after the presentations will be invited to submit their extended work for peer review and possible publication in the Special Issue of the Journal of Operational Research Society on Systems to Build Systems Organisers: Prof Edmund Burke Dr Ender Ozcan Dr Jonathan Thompson Purpose, aims and objectives: The current state of the art in the development of search methodologies is focused around the human design of bespoke systems, which are specifically tailored to the problem solving environment in hand. A particularly challenging goal is to develop intelligent systems which are capable of automatically building (and/or selecting) new systems. A paradigm shift has emerged in search methodologies over the past few years. Instead of taking the short term approach of dealing with single problems, there is a growing body of work that aims to raise the level of generality by adopting the long term approach of providing a more broadly applicable solution. These types of "systems to build systems" have the potential to significantly reduce the resource cost that is often associated with developing bespoke systems, since the search process is automated through adaptation/learning. The main purpose of the workshop is to bring researchers and practitioners together from the computer science, artificial intelligence and operational research fields, in order to discuss new ideas, recent advances, methodologies, frameworks, challenges and opportunities within the context of systems to build systems. Topics on the systems to build systems (e.g. hyper-heuristics) include (but not limited to): - new methodologies to build (and/or select) systems - automated (computer aided) design of heuristics (e.g., by genetic programming, grammatical evolution, etc.) - adaptive and self-tuning methodologies to build systems (e.g. evolutionary algorithms, multi-meme algorithms, adaptive operator selection, dynamic algorithm portfolios, reactive search, and others) - machine learning techniques to support/build systems (e.g. learning classifier systems, data mining, reinforcement learning, neural networks, and others) - fuzzy systems to build systems - applications and new challenging domains, particularly real-world problems - foundational studies towards the understanding of systems to build systems - classifications or categorisations of methodologies - generality of the methodologies to build systems - scalability issues - issues in multi-objective, discrete and continuous methodologies to build systems - issues in comparison of systems to build systems over multiple problem domains Invited speakers: - Prof Graham Kendall, University of Nottingham "Hyper-heuristics: Past, Present and Future" - Prof Erwin Pesch, University of Siegen "Metaheuristics: New Perspectives and Real Applications" Important dates: - Submission: June 7, 2010 - Notification of acceptance: June 15, 2010 - Workshop: July 18, 2010