Dr. Diarmuid O Donoghue

Homepage
Room 2.121 Callan Building,
National University of Ireland, Maynooth,
Maynooth,
Co.Kildare,
Ireland,
Email:
Phone: +353 1 7083851

Research Profile

Diarmuid O'Donoghue's research is focused on Evolutionary Algorithms for Optimisation under Constraints. We are focused on the use of Genetic Repair operators as a means of enforcing constrints upon the search space. We are evaluating the effectiveness of various genetic repair operators to enforce constrains upon a search space. He is particularly intrigued by the genetic repair process proposed in the plant Arabidopsis thaliana, and is apparent use of genetic information that originates in the grandparent generation - effectively by-passing the parent generation. As well as providing more robust constraint solving algorithms, this work it helping identify the conditions under which (deep) ancetral repair strategies might outperform more recent ancestral strategies.

Another of Diarmuid's interests is in computational models of how people re-use old solutions to solve new (similar) problems. Information is represented in the form of coloured graphs, with the solutions centring on finding a mappings between the old and new graphs. Analogical comparisons, spatial and geo-spatial analogies, isomorphic structure mapping, structure matching and k-edge + j-node coloured graphs. Topographic maps are effectively seen as irregular cellular automata.

Recent Publications

• Amy FitzGerald, Diarmuid O'Donoghue Genetic Repair Strategies inspired by Arabidopsis thaliana Proceedings of 20th Conference on Artificial Intelligence and Cognitive Science, August 200, UCD, Ireland. 58-67 2009
• Amy FitzGerald, Diarmuid O'Donoghue Investigating the Influence of Population and Generation Size on GeneRepair Templates China-Ireland International Conference on Information and Communications Technology (CIITC), August 2009, NUI Maynooth, Ireland. 2009
• Amy FitzGerald and Diarmuid P. O’Donoghue Genetic Repair for Optimization under Constraints Inspired by <i>Arabidopsis thaliana</i> Parallel Problem Solving from Nature, LNCS 5199, Springer Berlin & Heidelberg, Germany. ISBN 978-3-540-87699-1. 399-408 2008
• Diarmuid P. O'Donoghue Statistical Evaluation of Process-Centric Computational Creativity: Evaluating Computer-Generated Analogies 4th International Joint Workshop on Computational Creativity (IJWCC), Goldsmiths, University of London. 17-19 June. 2007
• O’Donoghue, Diarmuid. P. Mullally, Emma-Claire Extending Irregular Cellular Automata with Geometric Proportional Analogies GISRUK Conference, NUI Maynooth, Ireland. 11th - 13th April 353-358 2007
• Diarmuid P. O’ Donoghue , Amy Bohan, Mark T. Keane, Seeing Things: Inventive Reasoning with Geometric Analogies and Topographic Maps, New Generation Computing, 24 3 267-288, 2006
• Mullally, Emma-Claire. O’Donoghue, Diarmuid. P. Spatial Inference with Geometric Proportional Analogies Artificial Intelligence Review (Springer) 26 1-2 129-140 2006
• Mullally, Emma-Claire, D. O’Donoghue, ‘Geometric Proportional Analogies In Topographic Maps: Theory and Application’, Proceedings of 25th SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence, Cambridge, UK. Ann Macintosh, Richard Ellis, Tony Allen (Eds.) ISBN: 978-1-84628-223-2, pp 95-108, UK (December 12-14 2005). 0 2005
• D. O’Donoghue, A. Loughlin, A. Bohan, M. T. Keane, ‘Geometric Analogies and Topographic Maps’, Computational Creativity Workshop 2005, at IJCAI, (Eds.) P. Gervais, T. Veale, and A. Pease, Edinburgh, Scotland, pp 30-37, (July 30 2005). 0 2005
• D. Cleary, B. Danev, D. O’Donoghue, ‘Using Ontologies to Simplify Wireless Network Configuration’, Formal Ontologies Meet Industry, Verona, Italy, (June 9-10 2005). 0 2005

Computer Science Department , NUI Maynooth
Tel: +353-1-708 3847 | Fax: +353-1-708 3848| Email: admin@cs.nuim.ie