Nicolas Bousquet - Sorbonne Université - EDF Lab
Nicolas Bousquet - Sailing for science

Nicolas Bousquet

Head of Artificial Intelligence and Simulation at EDF Lab
Associate Professor at Sorbonne Université
Team "Statistics, Data, Algorithms"

Head of Statistics and Environment Group at SFdS
Coordinator of Risk and Uncertainty Working Group at GDR Mascot-Num

Contact

Desk (SU) 15-25.203

nicolas.bousquet 'at' sorbonne-universite.fr

nicolas.bousquet 'at' edf.fr

Research Gate / LinkedIn

Interests

Bayesian modeling, treatment of uncertainties, machine /deep learning, artificial intelligence
Risk analysis, health, industrial and environmental resource management, natural hazards



Course

January-March, 2020

Course M2 / ISUP Modélisation et Statistique Bayésienne Computationnelle

( Bayesian Modeling and Computational Statistics)

Research papers

N. Benoumechiara, B. Michel, P. Saint-Pierre, N. Bousquet (2019). Detecting and modeling worst-case dependence structures between random inputs of computational reliability models

N. Thiebaut, A. Simoulin, K. Neuberger, I. Ibnouhsein, N. Bousquet, N. Reix, S. Molière, C. Mathelin (2019). An innovative solution for breast cancer textual big data analysis

N. Bousquet, E. Chassot, E. Dortel, J. Million, A. Fonteneau, J.-P. Hallier (2019). A Bayesian Brownie-Petersen model for assessing the mortality and abundance of Indian Ocean tunas. Application to skipjack (Katswonus pelamis). In revision.

C. Sonigo, S. Jankowski, O. Yoo, O. Trassard, N. Bousquet, M. Grynberg, I. Beau, N. Binart (2018). High-throughput ovarian follicle counting by an innovative deep learning approach. Scientific Reports, 8.

C. Sonigo, S. Jankowski, O. Yoo, O. Trassard, N. Bousquet, M. Grynberg, I. Beau, N. Binart (2018). Comptage des follicules primordiaux par deep learning : l'intelligence artificielle au service de l'étude de la reproduction. Annales d'Endocrinologie, 79, 225.

N. Bousquet, T. Klein, V. Moutoussamy (2018). Approximation of limit state surfaces in monotonic Monte Carlo settings, with applications to classification. SIAM Journal of Uncertainty Quantification, 6: 1-33.

N. Bousquet (2018). Modeling extreme events in energy companies. In: Statistics Reference Online (Wiley Stats Ref), DOI:10.1002/9781118445112.stat08011

S. Fu, M. Couplet, N. Bousquet (2017). An adaptive kriging method for solving nonlinear inverse statistical problems. Environmetrics, 28 (4).

N. Pérot, N. Bousquet (2017). Functional Weibull-based models of steel fracture toughness for structural risk analysis: estimation and selection. Reliability Engineering and System Safety, 165: 355-367

L.J. Wolfson, N. Bousquet (2016). Elicitation. In: Statistics Reference Online (Wiley Stats Ref), DOI:10.1002/9781118445112.stat00231.pub2

M. Keller, A.-L. Popelin, N. Bousquet, E. Remy (2015). Nonparametric estimation of the probability of detection of flaws in an industrial component, from destructive and nondestructive testing data using Approximate Bayesian Computation. Risk Analysis, 35:1595-1610

A. Pasanisi, C. Roero, E. Remy, N. Bousquet (2015). On the practical interest of discrete Inverse Polya and Weibull-1 models in industrial reliability studies. Quality and Reliability Engineering International 31:1161-1175

S. Fu, G. Celeux, N. Bousquet, M. Couplet (2015). Bayesian inference for inverse problems occuring in uncertainty analysis, International Journal for Uncertainty Quantification, 5(1):73-98

N. Bousquet, M. Fouladirad, A. Grall, C. Paroissin (2015). Bayesian gamma processes for optimizing condition-based maintenance under uncertainty, Applied Stochastic Models in Business and Industry, 31(3): 360-379

P. Lemaître, E. Sergienko, A. Arnaud, N. Bousquet, F. Gamboa, B. Iooss (2015) Density modification based reliability sensitivity indices, Journal of Statistical Computation and Simulation, 85:1200-1223

E. Dortel, F. Sardenne, N. Bousquet, E. Rivot, J. Million, G. Le Croizier, E. Chassot (2015). An integrated Bayesian modelling approach for the growth of Indian Ocean yellow fin tuna Fisheries Research, 163: 69-84

N. Bousquet, E. Chassot, D. Duplisea, M. Hammill (2014). Forecasting the major influences of predation and environment on cod recovery in the northern Gulf of St. Lawrence. PloS ONE 9(2): e82836

B. Archambault, O. Lepape, N. Bousquet, E. Rivot (2014). Density dependence can be revealed by modeling the variance in the stock-recruitment process: an application to flatfishes. ICES Journal of Marine Science, doi:10.1093/icejms/fst203

E. Dortel, F. Massot-Granier, E. Rivot, J. Million, J.-P. Hallier, E. Morize, J.-M. Munaron, N. Bousquet, E. Chassot (2013). Accounting for Age Uncertainty in Growth Modeling, the Case Study of Yellowfin Tuna (Thunnus albacares) of the Indian Ocean.PloS ONE 8(4): e60886

N. Bousquet (2012). Accelerated Monte Carlo estimation of exceedance probabilities under monotonicity constraints. Annales de la Faculté des Sciences de Toulouse, 21(3): 557-591 (arxiv)

A. Pasanisi, S. Fu, N. Bousquet (2012). Estimating discrete Markov models from various incomplete data schemes. Computational Statistics & Data Analysis 56(9):2609-2625

N. Bousquet (2010). Eliciting vague but proper maximal entropy priors in Bayesian experiments. Stat. Papers. 51: 613-62

N. Bousquet, N. Cadigan, T. Duchesne, L.-P. Rivest (2010). Detecting and correcting underreported catches in fish stock assessment : trial of a new method. Canadian Journal of Fisheries and Aquatic Sciences.67(8): 1247-1261

E. Chassot, D. Duplisea, M. Hammill, A. Caskenette, N. Bousquet, Y. Lambert, G. Stenson (2009). The role of predation by harp seals (Phoca groenlandica) in the collapse and non-recovery of northen Gulf of St. Lawrence cod (Gadus morhua). Marine Ecology Progress Series, 379: 279-297

N. Bousquet (2008). Diagnostics of prior-data agreement in applied Bayesian analysis. Journal of Applied Statistics, 35:1011-1029

N. Bousquet,T. Duchesne, L.-P. Rivest (2008). Redefining the maximum sustainable yield for the Schaefer population model including multiplicativeenvironmental noise Journal of Theoretical Biology,254: 65-7

H. Bertholon, N. Bousquet, G. Celeux (2006). An alternative competing risk model to the Weibull distribution in lifetime data analysis. Lifetime Data Analysis, 12: 481-504

N. Bousquet (2005). Eliciting prior distributions for Weibull inference in anindustrial context. Communications in Dependability and Quality Management, 8: 12-19.

Books


Forthcoming in 2020:
Statistical Extreme Value Theory, with Applications to Natural Hazard Engineering
Springer: Heidelberg.

AI Report

Coordinator of


Design by Takuya

Book translation


Some details here about this book

Book chapter

Considérations décisionnelles pour la construction d'un ouvrage de protection contre les crues (with E. Parent, J. Bernier, A. Pasanisi, M. Keller). In: Approches Statistiques du Risque.Technip, 2014 (link - slides)

Patents

N. Pérot, N. Bousquet, M. Marques(2015). Method for determining the strength distribution and the ductile-brittle transition temperature of a steel product subjected to thermal variations. European Patent Deposit WO-2015-165962

E. De Oliveira, D. Vautrin, N. Bousquet, N. Paul, K. Zoubert-Ousseni (2017). Managing water-supply pumping for an electric production plant circuit. European Patent Deposit WO-2017-86603

Conference papers

N. Bousquet, F. Corset (2015). Exploring asymptotics of the MLE of imperfect repair ARA1 models for single data trajectories. MMR Congress, Tokyo

J. Bect, N. Bousquet, B. Ioosset al. (2014). Uncertainty quantification and reduction for the monotonicity properties of expensive-to-evaluate computer models. UCM 2014, Sheffield.

N. Bousquet, F. Douard (2014). Analyse bayésienne d'intensités de défaillance pour les études de gestion d'actif. Lambda-Mu 19, Dijon. Best Presentation Award

J. Bect, N. Bousquet, B. Ioosset al. (2014). Quantification et réduction de l’incertitude concernant les propriétés de monotonie d’un code de calcul coûteux à évaluer. SFdScongress, Rennes

M. Fouladirad,C. Paroissin, N. Bousquet, A. Grall (2013). Bayesian optimization of condition-based maintenance under uncertainty.MMR Proceedings, Stellenbosch.

V. Moutoussamy,N. Bousquet, B. Iooss, P. Rochet, T. Klein, F. Gamboa (2013). Comparing conservative estimations of failure probabilities using sequential designs of experiments in monotone frameworks.ICOSSAR, NY

E. Dortel,F. Sardenne, G. Le Croizier, N. Bousquet, E. Chassot (2012). Anintegrated Bayesian hierarchical model for growth of Indian Ocean yellowfin tuna. Indian Ocean Tagging Symposium,Mauritius.

E. Chassot,L. Dubroca, N. Bousquet, E. Dortel, S. Bonhommeau (2012). Two-stanza growth for tropical tunas: myth or reality? IndianOcean Tagging Symposium, Mauritius.

N. Bousquet,E. Chassot, E. Dortel, J. Million, J.P. Eveson, J.-P. Hallier, A. Fonteneau (2012). Preliminary assessments of tuna mortality rates from a Bayesian Brownie-Petersen model. Indian Ocean Tagging Symposium.

A.-L. Popelin,R. Sueur, N. Bousquet (2012). Encadrement et estimation de probabilités de défaillance dans un cadre monotone d'analyse de fiabilité structurale. Lambda-Mu 18,Tours, France.

M. Keller,N. Bousquet (2012).Estimation of a flaw size distribution from a data mixture based on destructive tests and non-destructive in-service inspections.ESREL-PSAM congress, Helsinki, Finland.

S. Fu, M. Couplet, N. Bousquet (2011). An adaptive kriging method for characterizing uncertainty in inverse problems. ISI congress, Dublin, Ireland

N. Bousquet(2011). Encadrement et estimation parcimonieuse de probabilités de dépassement en sortie d’un code de calcul monotone. SFdS congress, Tunis,Tunisia

N. Bousquet(2011). Calculating failure probabilities through constraintedMonte Carlo acceleration methods. MMR congress, Beijing, China

M. Keller, A. Pasanisi, E. Parent, N. Bousquet (2010). Bayesian and frequentist parametric prediction of a tail probability in an industrial reliability context. ISBA congress (poster), Benidorm, Spain.

A. Pasanisi,S. Fu, N. Bousquet (2010). Estimation de modèles markoviens discrets dans un cadre industriel fiabiliste à données manquantes. SFdS congress, Marseille, France.

A. Pasanisi, E. de Rocquigny, E. Parent, N. Bousquet (2009). Some useful features of the Bayesian setting while dealing with uncertainties in industrial practice. ESREL 09, Prague,Czech Republic

N. Bousquet(2009). CalibratingWeibull priors using virtual data. MMR congress, Moscow,Russia

N. Bousquet, T. Duchesne, L.-P. Rivest (2008) . Definition andestimation of biological reference points for halieutic resourcemanagement in stochastic frameworks, SSC-SFdS congress, Ottawa,Canada

N. Bousquet,G. Celeux (2006). Measures of Bayesian discrepancy between prior beliefs and data knowledge. ESREL 06, Lisbon, Portugal + Poster at ISBA congress (Benidorm, Spain)

N. Bousquet,G. Celeux, F. Billy, E. Remy (2006). Notions et mesures de cohérence bayésienne entre connaissance a priori et données observées. Lambda-Mu 15, Lille, France

N. Bousquet,G. Celeux, F. Billy, E. Remy (2006). Inférence des paramètres d'une loi de Weibull - Approches classique et bayésienne. Lambda-Mu 15, Lille, France

F. Josse, N.Bousquet, G. Celeux (2006). Vraisemblance d'enchaînements causaux: validation d'un explication a priori confrontée au retour d'expérience. Lambda-Mu 15, Lille, France

F. Billy,N. Bousquet, G. Celeux (2005). Modelling and eliciting expert knowledge with fictitious data in: Proceedings of the Workshop on the use of Expert Judgement for decision-making, CEA Cadarache

N. Bousquet,G. Celeux, E. Remy (2005). A protocol for integrating FED and expert data in a study of durability in: Proceedings of the Workshop on the use of Expert Judgement for decision-making, CEA Cadarache

N. Bousquet (2005). Choosing prior distributions for Weibull inference in a durability context: some propositions,International Symposium on Stochastic Models in Reliability, Safety, Security and Logistics, Israel

H. Bertholon,N. Bousquet, G. Celeux (2004). Un modèle de durée de vie à risques concurrents. SFdS congress, Montpellier, France + Poster at MMR congress (Sante Fe, USA)

Vulgarization

Avantages et enjeux de l'analyse statistique bayésienne en durée de vie industrielle. Lettre Techniques de l'Ingénieur : Risques Industriels, No 23, mars-avril 2007.

Mathématiques de la Planète Terre (2013) : Pour une pêche maximale durable / Quelle hauteur pour la digue ?


Read this article from Sciences au Sud (May-June-July 2015) about our work with Emmanuel Chassot and many other people about Indian Ocean tunas

A contribution within this book edited by Fanny Agostini , with contributions by François Gourand, Audrey Hasson, François Gabart, Pierre-Yves Cousteau, Nicolas Hulot and others :






In the past, I gave numerous talks about animal tagging campaigns, design of experiments and "real-life" statistics

Some preprints & research reports

Reference priors for nuisance parameters of Bayesian sequential population models

Monte Carlo acceleration of failure probabilities using monotone computer codes

Bayesian inference for inverse problems occuring in uncertainty analysis. INRIA RR-7995

A Bayesian analysis of industrial lifetime data with Weibull distributions. INRIA RR-6025

Subjective Bayesian statistics: agreement between prior and data. INRIARR-5900

An alternative competing risk model to the Weibull distribution in lifetime data analysis. INRIA RR-5265

Projects

IMT Project (Coordinator) METADEB

PEPS-AMIES Project (Coordinator) LYMPHOMATH

ANR Project (Contributor) AMMSI

ANR Project (Contributor) EMOTION

European Project (Advisor) CoForTips (FP7 ERA-NET BioDiverSa2)

European Project (Work Package Leader) MATCHING (H2020)

Some recent invited lectures

Statistical approaches can defeat machine learning rules. FINS Workshop, Agistri, 2020

About the geometry of black-box models. Academia Sinica, Taipei, 2019

High-dimensional copulas for solving unbalanced classification problems. GDRR Symposium, Madrid, 2017

Expert elicitation methods. ETICS Summer School, Porquerolles, 2017

Convexity and monotonicity in accelerated Monte Carlo methods. ALT Conference, Troyes, 2016

Quelques principes de modélisation bayésienne en traitement des incertitudes. CEA/DAM, Bruyères-Le-Châtel, 2016

Ph.D. co-advising

ShuaiFu (2009-2012) - Vincent Moutoussamy (2012-2015) - Nazih Benoumechiara (2015-2019) - Antoine Simoulin (2018-2021)

Ph.D. thesis

Analyse bayésienne de la durée de vie de composants industriels
(Elements of Bayesian lifetime analysis of industrial components)

Defended on 2006, at Paris XI (Orsay) University. Supervised by Gilles Celeux and Jean-Michel Marin.

Software tools

The QUANFRE project: a professional engineering software developed for EDF in collaboration with NOEO, incorporating frequentist and Bayesian methods for the computation of statistical decision-helping tools for lifetime analysis. The packaging software is written in C and and uses the .NET technology

The SIMCAB-Bayes project: R/C tools with Excel interfaces for Windows OS for frequentist and Bayesian computation of a prey-predator Bayesian model (first version is described here). This is the evolving result of various studies led with biologists from Department Fisheries and Oceans Canada and the Institut de Recherche pour le Développement.

R package MISTRAL (Methods In STructural Reliability AnaLysis), by V. Moutoussamy, B. Iooss and other contributors

Short bio

Graduated from ENSIMAG (2003)
Master of research in applied mathematics (Joseph Fourier University, Grenoble, 2003)
Ph. D. in Mathematics (Paris XI University / INRIA Saclay, Orsay, 2006), Select team
Post-doctoral internship at Laval University, Québec, Canada (2007-2008) with Thierry Duchesne and Louis-Paul Rivest

EDF Lab: Research engineer (2008-2013), expert researcher (2013-2017), R&D project manager (2016-2017)
Quantmetry: Scientific Director (2017-2020)

Miscellaneous (in French)

Terres amères (Prix Arthur-Rimbaud), Maison de Poésie, 2001.

Frondaisons (Prix Ecrits en Cité), Bacchanales, 2003.

Textes divers, Rimbaud Vivant & Le Coin de Table, 2003-2005.

Other miscellaneous (family, friends et al.)

Le studio Takuya (Instagram) qui a réalisé cette page et qui est spécialisé en pochettes de disques, livres et illustrations.

Le Monde d'Ida association qui gère la radio sans pub idaradio.fr (que je me permets de vous recommander).

La radio de l'open space (et d'un ami).

PostPoète
Lady Chatterton
Captain Americano

Gush

Roland Guarrigue
Divine Machine
Spinalien
Itinéraires polaires