Frédérick GARCIA
De WikiZ.
I am a Research director at INRA and a member of the Mathematics and Computer Science department in the Biometrics and Artificial Intelligence laboratory in Toulouse. I am head of the research group "Modélisation des Agroécosystèmes et Décision" (MAD).
Sommaire |
Research interests
Following AI (artificial intelligence) approaches, my research aims at developping methods for modelling, simulating and optimizing intelligent agents. More specifically, my research interests are in the following areas of AI:
- Planning under uncertainty, Markov decision problems
- Reinforcement learning
- Simulation Optimization
- Intelligent agents
- DEVS modelling and simulation
For the last ten years, the main application domain for my research in AI has been the management of agroecosystems, that can be seen as complex controlled dynamical systems. This domain is gaining interest today within the computer science community, with the recent launch of the NSF Expedition Computational Sustainability : Computational Methods for a Sustainable Environment, Economy, and Society.
Short bio
I hold a Ph.D degree in Artificial Intelligence from the Ecole Nationale Supérieure de l'Aéronautique et de l'Espace of Toulouse (Supaero, 1993) and an Engineer degree from the same school (1989). I got in 2004 a habilitation degree in Computer Science from the University Paul Sabatier (UPS) of Toulouse.
Collaborations
The MAD team (INRA-BIA, Toulouse), P. Fabiani (ONERA Toulouse), E. Jacopin (CREC Saint-Cyr), J.-E. Bergez (INRA-AGIR, Toulouse), L. O. Eriksson (SLU, Umea, Sweden), A. Reynaud (School of Economics, INRA-LERNA, Toulouse).
Teaching
- Supaero (Toulouse), Décision dans l'incertain, MDPs, 15h/year (2002-)
- Supaero (Toulouse), Décision dans l'incertain, Apprentissage par renforcement, 15h/year (2002-)
- NOVA PhD Course (SLU-Umea, Sweden). Markov Decision Processes, 20h (2008)
- Master 2 Biostatistique et Modélisation, (UPS, Toulouse). Processus Décisionnels de Markov, 15h/year (2006- )
- Master 1 Agrofood Chain, (UPS - Agricampus, Toulouse). Operations Research, 6h/year (2005-2006)
- Member of the Réflexives team as a <<facilitator>> (2004-2008)
Recent publications
N. Forsell, P. Wikström, F. Garcia, R. Sabbadin, K. Blennow and L. O. Eriksson. Management of the risk of wind damage in forestry: a graph-based markov decision process approach. Annals of Operations Research, 2009.
O. Crespo, J.-E. Bergez and F. Garcia. P2 hierarchical decomposition procedure: application to irrigation strategies design. Operational Research: An International Journal, 2009.
O. Crespo, J.-E. Bergez, and F Garcia. P2q hierarchical decomposition algorithm for quantile optimization: application to irrigation strategies design. Annals of Operations Research, 2009.
J.-E. Bergez, N. Colbach, O. Crespo, F. Garcia, M.-H. Jeuffroy, E. Justes, C. Loyce, N. Munier-Jolain and W. Sadok. Designing crop management systems by simulation. European Journal of Agronomy, 31, 2009.
J.-E. Bergez, P. Chabrier, F. Garcia, C. Gary, D. Makowski, G. Quesnel, E. Ramat, H. Raynal, N. Rousse and D. Wallach. Record: a new software platform to model and simulate cropping systems. In Proceedings of Farming Systems Design, 2009.
E. Rachelson, P. Fabiani and F. Garcia : Adapting an MDP planner to time-dependency: case study on a UAV coordination. In 4th Workshop ICAPS on Planning and Plan Execution for Real-World Systems: Principles and Practices for Planning in Execution, Thessaloniki, Greece, September 19-23rd 2009.
E. Rachelson, P. Fabiani and F. Garcia : TiMDPpoly : an improved method for solving timedependent MDPs. In 21st International Conference on Tools with Artificial Intelligence, Newark New Jersey, USA, November 2-5 2009.
E. Rachelson, F. Garcia and P. Fabiani: Extending the Bellman equation for MDPs to continuous actions and continuous time in the discounted case. In 10th International Symposium on Artificial Intelligence and Mathematics, Fort Lauderdale, Florida, USA, 2008.
E. Rachelson, G. Quesnel, F. Garcia and P. Fabiani : A simulation-based approach for solving generalized semi-Markov decision processes. In European Conference on Artificial Intelligence, Patras, Grèce, July 2008.
F. Garcia. Processus décisionnels de Markov en intelligence artificielle, volume 1. Chapitre 1 : Processus décisionnels de Markov. Sigaud O. and Buffet 0., Eds, Hermes Science Publishing, 2008.
O. Sigaud and F. Garcia. Processus décisionnels de Markov en intelligence artificielle, volume 1. Chapitre 2 : Apprentissage par renforcement. Sigaud O. and Buffet 0., Eds, Hermes Science Publishing, 2008.
L. Péret et F. Garcia. Processus décisionnels de Markov en intelligence artificielle, volume 2. Chapitre 5 : Approches de résolution en ligne. Sigaud O. and Buffet 0., Eds, Hermes Science Publishing, 2008.
Hobbies :-)
- Randomness, Quantum physics and other odd phenomena
- Drums (timbales, jazz and latin jazz)
Contact informations
- INRA Toulouse - BIA unit (office BIA #44), chemin de Borde-Rouge - BP 52627, 31326 Castanet-Tolosan Cedex, France.
- Tel: +33(0)5.61.28.52.83
- Mail: fgarcia at toulouse.inra.fr
Links
- RECORD project at MIA Toulouse
- VLE project at MIA Toulouse
- El Remedio on myspace
