INRIA et TAO for European projects

INRIA, the French national institute for research in computer science and control, is the only French public institute entirely dedicated to research in information and communication science and technology (ICST). Throughout its eight research units located in different regions, INRIA has a workforce of 4,350, 2,950 of whom are scientists from INRIA or from INRIA’s partner organizations such as CNRS (the French National Center for Scientific Research), Universities and leading Engineering Schools. INRIA has an annual budget of 265million euros, 40% of which comes from its own research contracts and licenses. INRIA plays a leading role in the following fields: “networks, telecoms and multimedia”, “complex systems and software” and "modeling, simulation and visualization". As its strategy closely combines scientific excellence with technology transfer, it develops collaborations with the economic world through strategic industrial partners and by creating companies (more than 100 start-ups in 25 years) - particularly through its subsidiary INRIA-Transfert, promoter of four start-up funds. INRIA was partner in about 120 FP6 projects (90 of them in the IST priority) and 186 (up to now) FP7 projects, including 30 ERC grants.

The TAO team is concerned with the cross-fertilization of Machine Learning, Data Mining and Evolutionary Computation. TAO was founded in September 2003, including researchers from the Centre National de la Recherche Scientifique, Institut National de Recherche en Informatique et Automatique and Laboratoire de Recherche en Informatique de l'Université Paris-Sud. Co-headed by Michèle Sebag (DR1 CNRS, member of PASCAL NoE Executive Committee) and Marc Schoenauer (DR1 INRIA, member of ACM-SIGEVO Executive Board), TAO is or was partner to numerous projects related to Learning and Stochastic Optimization: EvoTest (FP6-2006-IST-33742) and GENNETEC (FP6-IST-2006-34952), TRAVESTI (ANR-08-SYSC-017), SYMBRION (FP7-FET-IP in Proactive Initiative “Pervasive Adaptation”), DESCARWIN (ANR-09-COSI-002).

INRIA et TAO - version courte

INRIA is the only French public research institute entirely dedicated to information and communication science and technology. With a workforce of 4,350 including 2,950 scientists, INRIA has an annual budget of 265million euros, 40% of which comes from its own research contracts. INRIA plays a leading role in the field of "complex systems and software” and "modeling, simulation and visualization". INRIA was partner in about 120 FP6 projects (90 of them in the IST priority) and 186 (up to now) FP7 projects, including 30 ERC grants.

The TAO team is concerned with the cross-fertilization of Machine Learning, Data Mining and Evolutionary Computation. Co-headed by Michèle Sebag (DR1 CNRS, member of PASCAL NoE Executive Committee) and Marc Schoenauer (DR1 INRIA, member of ACM-SIGEVO Executive Board), TAO was partner to numerous projects related to Learning and Stochastic Optimization, including [here of course, pick up whichever project that fits the present proposal from the list in the long description]

TAO

The TAO Group is concerned with the cross-fertilisation of Machine Learning, Data Mining and Evolutionary Computation. TAO was founded in September 2003, including researchers from the Centre National de la Recherche Scientifique , Institut National de Recherche en Informatique et Automatique and Laboratoire de Recherche en Informatique de l'Université Paris-Sud.

TAO inherits the expertise in Evolutionary Computation from the Fractal Group in INRIA, and earlier from the EEAAX team (Artificial Evolution and Machine Learning Team) at Ecole Polytechnique.

TAO symmetrically inherits the expertise in Machine Learning and Data Mining from the Inference and Apprentissage (IA) Group at LRI-CNRS, which it largerly overlaps. IA, founded by Yves Kodratoff, ECCAI Fellow, was historically the first French Research Group in Machine Learning, participating in many national and European projects, among which: the ESPRIT Project Machine Learning Toolbox (MLT P2154), Inductive Logic Programming (ILP P6020; ILP2, LTR P20237)
, the Networks of Excellence in Machine Learning and in Inductive Logic Programming (MLNet P7115, ILPNet), and the European Science Foundation Project <i>Learning in Humans and Machines</i> (LHM).

TAO-IA currently participates into

• The Network of Excellence PASCAL, Pattern Analysis, Statistical Modelling and Computational Learning, continued as Pascal-2
• The Coordination Action ONCE-CS Open Network of Centres of Excellence in Complex Systems
• The Coordination Action KD-Ubiq Knowledge Discovery in Ubiquitous Environments.

Co-headed by Marc Schoenauer (DR INRIA) and Michèle Sebag (DR CNRS), TAO includes 7 permanent members and 10 PhD students.
TAO is situated in the Laboratoire de Recherche en Informatique, among the most ancient Labs in Computer Science in France, which includes circa 150 permanent members and 80 PhD students, dispatched in 11 research groups.

Michèle Sebag

Michele Sebag graduated at Ecole Normale Supérieure in Paris in Maths; she received her PhD in Computer Science in 1990 and her Habilitation in 1997. She is with the CNRS, Centre National de la Recherche Scientifique, since 1991, senior researcher (Directeur de Recherche) since 2003.
Primarily grounded in applications for Numerical Engineering, her research interests include Relational Learning, Statistical Learning, Evolutionary Computation and Genetic Programming, and Complex Systems, with applications in Autonomic Computing.

She is on the Editorial Board of Genetic Programming and Evolvable Hardware; she was
on the Editorial Board of Machine Learning Journal (2001-2008), Knowledge and Information Systems (2003-2007) and was Associate Editor for IEEE Trans on Evolutionary Computation (1997 2003).

She is member of the PC for the major conferences in AI, Machine Learning and Evolutionary Computation. She was co-chair of ILP 2001, vice-chair of ICDM 2003, area chair of ECAI 2004, ICML 2005 and ECML-PKDD 2005, ICML 2008 and ECML/PKDD 2008.

She participates in the Steering Committee of the PASCAL Network of Excellence (2003-2007), continued as Pascal -2, and she is manager of the Pascal CHALLENGE Programme.

5 Selected recent papers

Select the ones that are best suited for the targeted audience

• Akrour, R., Schoenauer, M., and Sebag, M.. Preference-based Policy Learning. In D. Gunopulos et al., eds.: Proc. ECML 2011, LNCS 6911, pp 12-27, Springer Verlag 2011.
• Fialho, A. , Da Costa, L. , Schoenauer, M. and Sebag, M.. Analyzing Bandit-based Adaptive Operator Selection Mechanisms. In: Annals of Mathematics and Artificial Intelligence – Special Issue on Learning and Intelligent Optimization, Springer Netherlands. September 2010.
• Loshchilov, I. , Schoenauer, M. and Sebag, M.. Comparison-Based Optimizers Need Comparison-Based Surrogates. In R. Schaefer et al., eds.: "Parallel Problem Solving from Nature (PPSN XI)", Springer, LNCS, Vol. 6238 : p. 364-373. September 2010.
• Delarboulas, P. , Schoenauer, M. and Sebag, M.. Open-Ended Evolutionary Robotics: an Information Theoretic Approach. In R. Schaefer et al., eds.: "Parallel Problem Solving from Nature (PPSN XI)", Springer, LNCS, Vol. 6238 : p. 334-342. 2010.
• Furtlehner, C. , Sebag, M. and Xiangliang, Z.. Scaling Analysis of Affinity Propagation. In: Physical Review E: Statistical, Nonlinear, and Soft Matter Physics, Vol. 81 : p. 066102. 2010.
• Arnold, L. , Paugam-Moisy, H. and Sebag, M.. Unsupervised Layer-Wise Model Selection in Deep Neural Networks. In: "ECAI'2010, European Conference on Artificial Intelligence", IOS Press : p. 915-920. Lisbon. 2010.
• Gaudel, R. and Sebag, M.. Feature Selection as a one-player game. In: "Proceedings of the 27th Annual International Conference on Machine Learning (ICML 2010)" : p. 359–366. 2010.
• Zhang, X. , Germain-Renaud, C. and Sebag, M.. Adaptively Detecting Changes in Autonomic Grid Computing. In: "Workshop on Autonomic Computational Science, in conjuction with 11th IEEE/ACM Grid Computing Conference", IEEE Computer Society : p. 387–392. 2010.
• A Multi-Objective Multi-Modal Optimization Approach for Mining Stable Spatio-Temporal Patterns. Michèle Sebag, Nicolas Tarrisson, Olivier Teytaud, Sylvain Baillet, Julien Lefevre, In Proc. IJCAI 2005, p. 859-864.
• Phase Transitions within Grammatical Inference. Antoine Cornuejols, Nicolas Pernot, Michele Sebag, In Proc. IJCAI 2005, p. 811-816.
• Fast Theta-Subsumption with Constraint Satisfaction Algorithms. J. Maloberti and M. Sebag. In Machine Learning Journal, 2004, Vol 55, p. 137-174.
• Ensemble Feature Ranking. K. Jong, J. Mary, A. Cornuejols, E. Marchiori, and M. Sebag. In Proc. Principles and Practice of Knowledge Discovery in Databases, ECML/PKDD 2004, Springer Verlag LNAI, p. 267-278.
• ROC-based Evolutionary Learning: Application to Medical Data Mining. M. Sebag, J. Azé? and N. Lucas Selected papers from Artificial Evolution'03, Springer Verlag LNCS, p. 384-396.

Marc Schoenauer

Marc Schoenauer graduated at Ecole Normale Supèrieure in Paris, and passed a PhD thesis in Numerical Analysis at Université Paris 6 in 1980. From 1980 until Aug. 2001 he has been full time researcher at CNRS (French National Research Center), working at CMAP (the Applied Maths Laboratory) at École Polytechnique. He then became Directeur de Recherche at INRIA and worked in the Projet Fractales before founding TAO.

Marc Schoenauer has been working in the field of Evolutionary Computation since the early 90s, is author of more than 60 papers in journals and major conferences of that field. He is or has been advisor of 18 PhD students.
He has also been part-time Associate Professor at Ecole Polytechnique in the Applied Maths Department from 1990 to 2004, was in charge of the Optimization track at Ecole Nationale des Ponts et Chaussées from 2001 to 2005 and is currently teaching the Master2 course in Evolutionary Computation and Robotics at Université Paris-Sud.

Marc Schoenauer is Editor in Chief of Evolutionary Computation Journal, has been associate editor of the IEEE Transactions on Evolutionary Computation from its start in 1996 until 2004, and is associate editor of the Genetic Programming and Evolvable Machines Journal, of the Journal of Applied Soft Computing and of the recently founded Theoretical Computer Science - Theory of Natural Computing (TCS-C). He was member of the Executive Committee of the European Network of Excellence on evolutionary computation (Evonet) since its first funding in 1996, has served in the IEEE Technical Committee on Evolutionary Computation from 1995 to 1999, and is member of the PPSN Steering Committee. He has been member of numerous program committees of international conferences, and was general chair of PPSN'2000 conference in Paris.

He is Senior Fellow, and member of the Board, of the late ISGEC (International Society of Genetic and Evolutionary Computation), that now became an ACM SIG - SIGEVO (Special Interest Group on Evolutionary Computation). He was the founding president of Evolution Artificielle, the French Society for Evolutionary Computation, who organizes the series of conferences Evolution Artificielle. He is on the board of AFIA - the French Society for Artificial Intelligence, from which he was president (2002-2004).

Some selected recent papers(!)

Select the ones that are best suited for the targetted audience

• Akrour, R., Schoenauer, M., and Sebag, M.. Preference-based Policy Learning. In D. Gunopulos et al., eds.: Proc. ECML 2011, LNCS 6911, pp 12-27, Springer Verlag 2011.
• Fialho, A. , Da Costa, L. , Schoenauer, M. and Sebag, M.. Analyzing Bandit-based Adaptive Operator Selection Mechanisms. In: Annals of Mathematics and Artificial Intelligence – Special Issue on Learning and Intelligent Optimization, Springer Netherlands. September 2010.
• Loshchilov, I. , Schoenauer, M. and Sebag, M.. Comparison-Based Optimizers Need Comparison-Based Surrogates. In R. Schaefer et al., eds.: "Parallel Problem Solving from Nature (PPSN XI)", Springer, LNCS, Vol. 6238 : p. 364-373. September 2010.
• Delarboulas, P. , Schoenauer, M. and Sebag, M.. Open-Ended Evolutionary Robotics: an Information Theoretic Approach. In R. Schaefer et al., eds.: "Parallel Problem Solving from Nature (PPSN XI)", Springer, LNCS, Vol. 6238 : p. 334-342. 2010.
• Nicolau, M. and Schoenauer, M.. On the Evolution of Scale-Free Topologies with a Gene Regulatory Network Model. In: BioSystems Journal, Vol. 98(3) : p. 137-148. December 2009.
• V. Pratap Singh, M. Schoenauer, M. Leger, A geologically-sound representation for evolutionary multi-objective subsurface identification, D. Corne, ed., CEC'2005, pp. 454--462, IEEE Press, 2005.
• Y. Semet and M. Schoenauer. An efficient memetic, permutation-based evolutionary algorithm for real-world train timetabling. In Proc. CEC'2005, IEEE Press, 2005.
• A. Auger, M. Schoenauer, O. Teytaud. Local and global Order 3/2 convergence of a Surrogate Evolutionary Algorithm. H.-G. Beyer et al., eds., GECCO'2005, ACM Press, 2005.
• C. Kavka, P. Roggero, and M. Schoenauer. Evolution of Voronoi based fuzzy recurrent controllers. H.-G. Beyer et al., eds., GECCO'2005, pp 1385--1392, ACM Press, 2005.
• N. Godzik and M. Schoenauer and M. Sebag, Robustness in the long run: Auto-teaching vs Anticipation in Evolutionary Robotics, in X. Yao et al., eds., Parallel Problem Solving from Nature VIII, pp 932-941, LNCS 3242, Springer Verlag, 2004.