Sylvain GELLY's Home Page

Introduction

Publications

MoGo

OpenDP

Teaching

Fun

Note that the list is neither complete nor really up to date.

PhD thesis

National/International Journals

Sylvain Gelly, Olivier Teytaud
Bayesian networks : a better than frequentist approach for parametrization, and a more accurate structural complexity measure than the number of parameters .
RIA (Revue d'Intelligence Artificielle) http://ria.revuesonline.com/
To appear in english (2006) (p717-755)
Download (draft) pdf version

International Publications

Sylvain Gelly, David Silver
Achieving Master Level Play in 9 x 9 Computer Go.
Proceedings of the Twenty-Third AAAI Conference on Artificial Intelligence, AAAI 2008, Chicago, Illinois, USA, July 13-17, 2008, p. 1537-1540
Link to the pdf on the editor site
Download pdf version

Sylvain Gelly, David Silver
Combining Online and Offline Knowledge in UCT.
International Conference of Machine Learning, ICML 2007, Corvallis Oregon USA, p. 273-280
Download pdf version

Sylvain Gelly, Yizao Wang, Remi Munos and Olivier Teytaud
Modifications of UCT with Patterns in Monte-Carlo Go.
Technical Report 6062, INRIA
Download pdf version

Sylvain Gelly and Yizao Wang
Exploration exploitation in Go: UCT for Monte-Carlo Go.
NIPS-2006, Online trading between exploration and exploitation, Whistler Canada, 8 December
Download pdf version

C. Hartland, S. Gelly, N. Baskiotis, O. Teytaud and M. Sebag
Multi-Armed Bandit, Dynamic Environments and Meta-Bandits .
NIPS-2006, Online trading between exploration and exploitation, Whistler Canada, 8 December
Download pdf version

Olivier Teytaud and Sylvain Gelly
General lower bounds for evolutionary algorithms.
PPSN-2006, Reykjavik, 9-13 September, Parallel Problem Solving from Nature
Download (long draft) pdf version

Sylvain Gelly, Jeremie Mary and Olivier Teytaud
On the ultimate convergence rates for isotropic algorithms and the best choices among various forms of isotropy.
PPSN-2006, Reykjavik, 9-13 September, Parallel Problem Solving from Nature
Download (long draft) pdf version

Sylvain Gelly, Sylvie Ruette and Olivier Teytaud
Comparison-based algorithms: worst-case optimality, optimality w.r.t a bayesian prior, the intraclass-variance minimization in EDA, and implementations with billiards.
PPSN-BTP-Workshop, PPSN-2006, Reykjavik, 9-13 September, Parallel Problem Solving from Nature
Download (draft) pdf version

Sylvain Gelly, Olivier Teytaud, and Christian Gagne
Resource-Aware Parameterizations of EDA.
IEEE CEC 2006, Vancouver, BC, Canada, July 16-21, Congress on Evolutionary Computation
Download (draft) pdf version

Sylvain Gelly, Jeremie Mary and Olivier Teytaud
Learning for stochastic dynamic programming.
ESANN 2006, 26 - 28 April, Bruges Belgium 11th European Symposium on Artificial Neural Networks
Download (draft) pdf version

S. Gelly, N. Bredeche, M. Sebag
From Factorial and Hierarchical HMM to Bayesian Network : A Representation Change Algorithm.
SARA 2005 : Symposium on Abstraction, Reformulation and Approximation (p107-120)
Link to the editor
Download (draft) pdf version

O. Teytaud, S. Gelly, N. Bredeche, M. Schoenauer
A Statistical Learning Theory Approach of Bloat.
GECCO 2005 (poster) Genetic and Evolutionary Computation Conference (p427-428)
Download (long) pdf version

Y. Semet, S. Gelly, M. Schoenauer, M. Sebag
Artificial Agents and Speculative Bubbles.
CF'04: International Conference on Computational Finance & its Applications
(PKDD-2004), Pisa, Italy, Sept. 20-24, 2004.
Download pdf version (154k)

Workshop / National Conferences

Sylvain Gelly, Jeremie Mary, Olivier Teytaud :
Taylor-based pseudo-metrics for random process fitting in dynamic programming : expected loss minimisation and risk management. Processus decisionnels de Markov et Intelligence Artificielle (PDM et IA), Lille juin 2005. p21-35.
Download pdf (184k)
S. Gelly, O. Teytaud
Statistical asymptotic and non-asymptotic consistency of bayesian networks : convergence to the right structure and consistent probability estimates.
CAP 2005 Conference Apprentissage Nice, France, du 31 mai au 3 juin 2005 p. 147-162
Download pdf version

O. Teytaud, S. Gelly, N. Bredeche, M. Schoenauer
Apprentissage statistique et programmation genetique: la croissance du code est-elle inevitable? (A Statistical Learning Theory Approach of Bloat).
CAP 2005 Conference Apprentissage Nice, France, du 31 mai au 3 juin 2005 p. 163-178
Download pdf version

S. Gelly, N. Bredeche, M. Sebag
HMM hierarchiques et factorises : mecanisme d'inference et apprentissage e partir de peu de donnees.
CAP 2005 (poster) Conference Apprentissage Nice, France, du 31 mai au 3 juin 2005 p. 143-144
Download (long) pdf version

Sylvain Gelly, Jeremie Mary, Olivier Teytaud :
Taylor-based pseudo-metrics for random process fitting in dynamic programming : expected loss minimisation and risk management. Nice, France, du 31 mai au 3 juin 2005 p. 183-184
Download pdf of extended abstract (80k)
Download pdf of poster (398k)

S. Gelly, N. Bredeche, M. Sebag
Inference dans les HMM hierarchiques et factorises : changement de representation vers le formalisme des Reseaux Bayesien.
EGC 2005 Atelier sur les Modeles Graphiques Probabilistes
Download ps version| Download pdf version
Download long version (ps)

Miscellaneous