Luis DA COSTA - Personal Website

Post-Doctoral Researcher at:
TAO Team-INRIA Saclay
Université Paris-Sud XI

Quality Manager and Researcher/Programmer
at the Evotest European Project

Main Collaborators:
Marc Schoenauer (INRIA Saclay)
Alvaro Fialho (Microsoft Research)
Tanja Voss (Universitat Politecnica de Valencia)

Contact:
Phone: (+1)(514)572.61.15
Email: dacosta (dot) le (at) gmail.com

My CV can be found here (PDF, 80 Kb) and more information at linkedin.

Main Topics of Interest

* Automatic parameter tuning of evolutionary algorithms;
* Optimization and Search techniques (evolutionary algorithms and other metaheuristics);
* Complex Systems and Agent-Based Simulation
* Machine Learning

Short Biography

Born in Montevideo, Uruguay, I have been on the road for most of my life.

In February 1994, I graduated in Computer Science from the Universidad Simon Bolivar, Caracas Venezuela. After that, I started a Master Degree in Computer Science, specialization in Artificial Intelligence, and taught several courses in the CS Departement. I finished my Master's degree in February 1998, after working in the AI Lab for my thesis. I worked in industry as a Software Project Manager at Eurobanco (Caracas) and I was one of the founders and Project Manager at Kynesoft Sistemas.

I moved to Montreal in 2001, and a new interest grew in me for the research world: I started a PhD at the Ecole de Technologie Superieure and graduated in December 2006; in April 2007, I started working with Marc Schoenauer as part of his TAO team in Saclay (south of Paris), France. I am working on Automatic Parameter Tuning for Evolutionary Algorithms; I am also applying evolutionary computation ideas to solving software engineering problems (specifically, testing), as part of the EvoTest European Project (of which I am the Quality Manager). As part of my research, I am also one of the main developers of GUIDE, a library with a Graphical User Interface that allows non specialists of Evolutionary Computation (EC) to quickly develop their own applications. It is an Open Source project, published here.

I am currently based in Montreal.

Publications

Journals and Lecture Notes (in Computer Science and Artificial Intelligence)

A. Fialho, Da Costa, Luis, M. Schoenauer and M. Sebag 2008. Extreme Value Based Adaptive Operator Selection. In Parallel Problem Solving from Nature PPSN X, volume 5199/2008, pages 175-184. Springer Berlin / Heidelberg, 2008.
Chion, Clement, Landry, J.A. and Da Costa, Luis E. 2008. A Genetic Programming Based Method for Hyperspectral Data Information Extraction: Agricultural Applications. IEEE Transactions on Geoscience and Remote Sensing Volume 46, Issue 8 (August). Pages: 2446-2457.
Da Costa, Luis E. Landry, J.A. and Levasseur, Yan 2007. Treating Noisy Data Sets with Relaxed Genetic Programming. In N. Monmarche et al. (Eds.): EA 2007, LNCS 4926, pp. 1-12, 2008. Springer-Verlag Berlin Heidelberg
Landry, J.A.; Da Costa, Luis E. and Bernier, T. 2006. Discriminant Feature Selection by Genetic Programming: Towards a Domain Independent Multi-Class Object Detection System. Journal of Systemics, Cybernetics and Informatics, 3(1).
Da Costa, Luis E. and Landry, J.A. 2003. On the Applicability of L-Systems and Iterated Functions Systems for Grammatical Synthesis of 3D Models. Lecture Notes in Artificial Intelligence 2671: 614-615 2003.

Peer-Reviewed Conferences

Da Costa, Luis and M. Schoenauer 2009. Bringing Evolutionary Computation to Industrial Applications with GUIDE. Proceedings of GECCO'09 (Montreal, Canada) (to appear).
A. Fialho, Da Costa, Luis, M. Schoenauer and M. Sebag 2009. Dynamic Multi-Armed Bandits and Extreme Value-based Rewards for Adaptive Operator Selection in Evolutionary Algorithms. LION'09: Electronic Proceedings of the 3rd Interna- tional Conference on Learning and Intelligent OptimizatioN (to appear).
Da Costa, Luis, A. Fialho, M. Schoenauer and M. Sebag 2008. Adaptive Operator se- lection with dynamic multi-armed bandits. Proceedings of GECCO 2008 (Atlanta, Georgia, USA), pp. 913{921.
Da Costa, Luis and M. Schoenauer 2007. GUIDE, a Graphical User Interface for Evolutionary Algorithms Design. GECCO Workshop on Open-Source Software for Applied Genetic and Evolutionary Computation (SoftGEC), J. H. Moore (editor)
Da Costa, Luis E. and Landry, J.A. 2006. Relaxed Genetic Programming. GECCO 2006. Seattle, WA, USA. July 8th-12th.
Chion, Clément, Da Costa, Luis E. and Landry, J.A. 2006. Genetic Programming for Agricultural Purposes. GECCO 2006. Seattle, WA, USA. July 8th-12th.
Da Costa, Luis E. and Landry, J.A. 2005. Generating grammatical plant models with genetic algorithms. Proceedings of the International Conference on Adaptive and Natural Computing Algorithms. Coimbra, Portugal. March 21-23. SpringerWien, New York. 230-234.
[pdf]
Landry, J.A., Da Costa, Luis E. and Bernier T., 2004. Discriminant Feature Selection by Genetic Programming: Towards a Domain Independent Multi-Class Object Detection System. SCI 2004 : The 8th World Multiconference on Systemics, Cybernetics and Informatics. Orlando, Florida, USA. July 18 - 21, 2004.
Best paper award on session Image and Multidimensional Signal Processing V
Da Costa, Luis E. and Landry, J.A. 2004. Synthesis of Fractal Models for Plants and Trees: First Results. Fractal 2004 : Complexity and Fractals in Nature. Vancouver, Canada. April 2004.
[pdf]
Da Costa, Luis E. and Landry, J.A. 2004. A Convivial Visualization Environment for LSystems. Fractal 2004 : Complexity and Fractals in Nature. Vancouver, Canada. April 2004. [pdf]
Da Costa, Luis E. and Landry, J.A. 2003. On the Applicability of L-Systems and Iterated Functions Systems for Grammatical Synthesis of 3D Models. AI'2003, Halifax, Canada. June, 2003. [pdf]

Others

* My page for the Genetic Programming Bibliography (maintained by Bill Langdon) is here

* I attended the 2005 Complex Systems Summer School of the Santa Fe Institute. Worked with Dalia Terhesiu in a model for studying the diffusion of ideas in a society (presentation here) and I studied the relationship between structural and dynamical complexities in networks.

* My CV can be found here (PDF, 80 Kb), ici aussi (PDF, 80 Kb), or even industrially here (PDF, 48 Kb).

This web-page is in construction (seriously!!). If you want to consult my old web page, just click here

Last Updated on July 1st., 2009.