Régis SABBADIN

De MIAT INRA
Aller à : navigation, rechercher


Régis Sabbadin has been a research scientist in the Applied Mathematics and Computer Science laboratory  in Toulouse, since 1999.

Brief bio

Diplomas

  • Engineer degree in aeronautics from the ENSICA school (now ISAE) in Toulouse in 1993
  • PhD degree in Computer Science from the University of Toulouse, on An ordinal approach to decision under uncertainty: Axiomatization, logical representation and application to sequential decision-making in 1998.
  • "Habilitation à diriger des Recherches" from the University of Toulouse, on Models and algorithms for sequential decision under uncertainty in 2009.

Research Domain

My research focuses on Artificial Intelligence methods for automated decision in sequential decision problems under uncertainty

For several years now, my research has been mainly focused on the theory of complex factored decision problems, with applications to agricultural and natural resource management problems, especially in epidemiology, agroforestry and wildlife preservation. The research issues are :

  • the control of spatio-temporal processes and
  • the design of sampling methods for spatial processes.

The methods developed for both research issues rely, more or less heavily, on the Markov Decion Processes framework.

In the past, I have been working on qualitative approaches to decision making, mainly based on the qualitative Possibility theory. I still may be working on this subject in the future, from time to time...

Research themes

  • Control of spatio-temporal processes
  • Design of sampling methods for spatial processes
  • Qualitative approaches to decision making

Selected recent publications

Around Markov decision processes

Factored MDP
  • Sabbadin (R.), Peyrard (N.) and Forsell (N.). A framework and a mean-field algorithm for the local control of spatial processes. International Journal of Approximate Reasoning, 53(1), 66-86, 2012.
Spatial Sampling
  • Bonneau (M.), Gaba (S.), Peyrard (N.) and Sabbadin (R.). Reinforcement learning-based design of sampling policies under cost constraints in Markov random fields: Application to weed map reconstruction. Computational Statistics and Data Analysis, 73, pp. 30-44, 2014.
  • Peyrard (N.), Sabbadin (R.), Spring (D.), Brook (B.), Mac Nally (R.). Model-based adaptive spatial sampling for occurrence map construction. Statistics and Computing, 23(1), pp. 29-42, 2013.
Other recent MDP-linked papers
  • Sabbadin (R.) and Viet (A.F.). A Tractable Leader-Follower MDP Model for Animal Disease Management. Proc. of AAAI 2013 -CompSust, 2013
  • Chades (I.), Carwardine (J.), Martin (T.G.), Nicol (S.), Sabbadin (R.) and Buffet (O.). MOMDPs: A Solution for Modelling Adaptive Management Problems. Proc. of AAAI 2012 -CompSust, 2012.

Environment and natural resources management

  • Tixier (P.), Peyrard (N.), Aubertot (J.N.), Gaba (S.), Radoszycki (N.), Caron-Lormier (G.), Vinatier (F.), Mollot (G.) and Sabbadin (R.). Modelling Interaction Networks for Enhanced Ecosystem Services in Agroecosystems. Ecological Networks in an Agricultural World. vol. 49, 437-480, Elsevier, 2013.
  • Forsell (N.), Wikström (P.), Garcia (F.), Sabbadin (R.), Blennow (K.) and Eriksson (L. O.). Management of the risk of wind damage in forestry: a graph-based Markov decision process approach. Annals of Operations Research, 190(1), 57-74, 2011.
  • Sabbadin (R.), Spring (D.) and Rabier (C.-E.). Dynamic reserve site selection under contagion risk of deforestation. Ecological Modelling, vol. 201, pp. 75--81, 2007.

Qualitative decision under uncertainty

  • Garcia (L.) and Sabbadin (R.). Complexity results and algorithms for possibilistic influence diagrams. Artificial Intelligence, vol. 172, pp. 1018-1044, 2008.
  • Fargier (H.) and Sabbadin (R.). Qualitative decision under uncertainty: back to expected utility. Artificial Intelligence, vol. 165, pp. 245--280, 2005.

For a more comprehensive list of publications.

Matlab Toolboxes

MDP Toolbox

A Matlab/Scilab/Octave/R toolbox grouping MDP solution utilities.

Authors: Chades I., Chapron G., Cros MJ., Garcia F., Sabbadin R.

 Paper: Chades I., Chapron G., Cros MJ., Garcia F., Sabbadin R. (2014). MDPtoolbox: a multi-platform toolbox to solve stochastic dynamic programming problems. Ecography 37:916-920.

GMDP Toolbox

Authors: Cros MJ, Peyrard N., Sabbadin R.

 Paper: Sabbadin (R.), Peyrard (N.) and Forsell (N.). A framework and a mean-field algorithm for the local control of spatial processes. International Journal of Approximate Reasoning, 53(1), 66-86, 2012.

Project participations

  • French ANR project AgroBiose (2014-2017).
  • French Ecophyto project VESPA (2013-2016).
  • European project PURE (2011-2015).
  • French ANR project LARDONS (2010-2014).
  • French ANR project FICOLOFO (2010-2014).

PhD and Master Students, Post-doc

PhD students

  • Julia Radoszycki (PhD), cosupervised with S. Gaba (INRA-SPE, Dijon) and N. Peyrard (INRA-MIA, Toulouse). Méthodes de conception par optimisation de stratégies spatiales de gestion de communautés de plantes adventices, (2012- ).
  • Mathieu BONNEAU (PhD), cosupervised with S. Gaba (INRA-SPE, Dijon) and N. Peyrard (INRA-MIA, Toulouse). Développement d'une méthode adaptative d'échantillonnage spatialisé pour caractériser la composition du stock semencier des espèces adventices à l'échelle de la mosaïque paysagère, (2009-2012).
  • Nicklas FORSELL (PhD) with F. Garcia (INRA-MIA, Toulouse), L.-O. Eriksson (SLU, Umea, Sweden) and P. Wikström (SLU, Umea, Sweden). MDP Approaches for Complex Forest Planning Under Uncertainty, (2005-2009).
  • Nasolo RAVOANJANAHARY (PhD) with J. Lang (CNRS-UPS, Toulouse). Processus décisionnels épistémiques et élicitation interactive de préférences, (2004-2008).

Recent Master students

  • Vincent COURJAULT-RADE (Master "Ingénierie Mathématique A Toulouse"), cosupervised with N. Peyrard. Optimal stopping appliqué à la gestion d’espèces sur graphe à structure incertaine, 2014
  • Julia RADOSZYCKI (INSA Toulouse, Master "Statistique et Probabilité"), cosupervised with N. Peyrard. Algorithme VBEM (Variational Bayesian EM) pour l’estimation du processus de Cox log gaussien, 2012.
  • Gabriel SIRVENT (Master "Intelligence artificielle, Robotique et Reconnaissance des formes",  cosupervised with N. Peyrard. Echantillonnage dans les Markov Logic Network, 2012.
  • Mathieu BONNEAU  (Master  "Ingénierie Mathématique A Toulouse", option statistiques), cosupervised with N. Peyrard. Echantillonnage spatial optimal basé sur la méthode du krigeage , 2009.
  • Usman FARROKH (Master "Informatique et Télécommunications, option intelligence artificielle), cosupervised with N. Peyrard. Echantillonnage spatial optimal pour la reconstruction de cartes d'occurrence, 2009.
  • Benjamin BORGY (Master "Biostatistiques et Modélisation"), cosupevised with J.-N. Aubertot and N. Peyrard. Exploitation du cadre des PDM sur graphe pour l'aide à la conception de stratégies de gestion collectives de résistances variétales au phoma du colza, 2008.

Post-Docs

  • Alexandre Albore, co-supervised with N. Peyrard and F. Teichteil (ONERA, toulouse). Design of mathematics and artificial intelligence tools for the mapping and sustainable management of crops pests using autonomous UAVs. 2014-2015.
  • Will Probert, co-supervised with E. McDonald-Madden (CEED, CSIRO, University of Queensland, Australia) and N. Peyrard. Computational issues surrounding the dynamic optimisation of management of an ecological food web. 2012 and 2013.
  • Alana Moore, co-supervised with M. MacCarthy (Univ. of Melbourne, Australia) and N. Peyrard. Control of partially observed, competitive contact processes on graphs: developing a new mathematical approach for managing invasive species or plant pathogens. 2010.


Teaching

I have been teaching regularly, between 20h and 50H per year since 2002. This teaching activity is linked to my research fields (operations research and decision under uncertainty). A significant part of this teaching activity has been directed to foreign students, and performed in english. From 2014 on, I have decided to stop (temporarily) my teaching activities.

Past teaching experience :

  • NOVA PhD Course (SLU-Umea, Sweden). Markov Decision Processes, 20h (2008).
  • Master 2 Biostatistique et Modélisation, (UPS, Toulouse). Processus Décisionnels de Markov, 15h/year (2005-2014).
  • Master 1 Agrofood Chain, (UPS - Agricampus, Toulouse). Operations Research, 6h/year (2005-2009).
  • Ecole Supérieure de Commerce de Rennes (Rennes). Operations Research, 40h/year (2002-2006).

What do I look like (on a rugby field)?

Regis1.jpg

Regis2.jpg

Regis3.jpg

Regis4.jpg


Génotoul BioInfo
Équipe RECORD
IMABS
Outils personnels