Joseph Rynkiewicz
Email : Joseph.Rynkiewicz@univ-paris1.fr
Postal Address :
SAMM, Université Paris 1
90, rue de Tolbiac - 75634 PARIS CEDEX 13 - FRANCE
Domaine de recherche
Réseaux de neurones
Séries temporelles
Cours
Deep learning
Quelques démonstrations de Deep-Learning avec Pytorch :
https://github.com/JosephRynkiewicz
Inégalités oracles pour le hold-out :
Analyse des données en M1
Statistique en M1
Séries temporelles en M1
Publications
Mixtures of nonlinear Poisson autoregressions, P. Doukhan, K. Fokianos, J. Rynkiewicz, Journal of Time Series Analysis (2020).
https://doi.org/10.1111/jtsa.12558
Spectral estimation for non-linear long range dependent discrete time trawl processes. P. Doukhan, F. Roueff, J. Rynkiewicz, Electronic Journal of Statistics, Vol. 14, No. 2, 3157-3191 (2020).
https://doi.org/10.1214/20-EJS1742
Asymptotic statistics for multilayer perceptron with ReLU hidden units. J. Rynkiewicz, Neurocomputing 342, p. 16-23 (2019)
A classe of random field memory models for mortality forecasting, P. Doukhan,, D. Pommeret, J. Rynkiewicz, Y. Sahli, Insurance : Mathematics and Economics, 77, p. 97-110 (2017).
Asymptotic for regression models under loss of identifiability, J. Rynkiewicz, Sankhya A, 78:2, p. 155-179 (2016).
Assessment of the influence of education level on voting intention for the Extreme right in France. J. Rynkiewicz, M. R. Benmakrelouf, W. Karouche, Revista investigacion operacional. 37:3 p. 205-215 (2016).
Estimating the number of regimes of non-linear autoregressive models. J. Rynkiewicz, Revista investigacion Operacional. 34:2 p. 117-127 (2013).
Neural networks for Complex Data. M. Cottrell, M. Oltéanu, F. Rossi, J. Rynkiewicz and N. Villa-Vialaneix, Künstliche Intelligenz. 26:4, p. 373-380 (2012).
Asymptotic properties of autoregressive regime-switching models, M. Oltéanu, J. Rynkiewicz, ESAIM : Probability and Statistics, 16, p. 25-47, (2012).
General bound of overfitting for MLP regression models., J. Rynkiewicz, Neurocomputing, 90, p. 106-110 (2012).
Asymptotic properties of mixture-of-experts models, M. Oltéanu, J. Rynkiewicz, Neurocomputing, 74 (9), p. 1444-1449 (2011).
Estimating the number of components in a mixture of multilayer perceptrons, M. Oltéanu, J. Rynkiewicz, Neurocomputing, 71 (7-9), p. 1321-1329 (2008).
A 24-h forecast of ozone peaks and exceedance levels using neural classifiers and weather predictions, A.-L. Dutot, J. Rynkiewicz, J. Rude, F. Steiner, Environmental Modelling and Software, 22 (9), p. 1261-1269 (2007).
Estimation and test for multi-dimensional regression models, J. Rynkiewicz, Communication in Statistics - Theory and Methods, 36 (14), p. 2655-2671 (2007).
Estimation consistante de l’architecture des perceptrons multicouches, J. Rynkiewicz, Comptes Rendus de l’Académie des Sciences - Series I - Mathematics, 342 (9), p. 697-700 (2006).
Efficient estimation of multidimensional regression models using multilayer perceptrons, J. Rynkiewicz, Neurocomputing, 69 (7-9), p. 671-678 (2006).
Self organizing map algorithm and distortion measure, J. Rynkiewicz, Neural Networks, 19 (6-7), p. 830-837 (2006).
Consistance d’un estimateur de minimum de variance étendue, J. Rynkiewicz, Comptes Rendus de l’Académie des Sciences - Series I - Mathematics, 341, p. 133-136 (2005).
Some known facts about financial data, M. Cottrell, E. De Bodt, J. Rynkiewicz, European Journal of Economic and Social Systems 17, 167-182 (2004).
Estimation of linear autoregressive models with Makov-switching. Rynkiewicz J., Investigacion Operacional, La Havane, Cuba. 25 :2 p. 166-173 (2004).
Caractérisation des crises financières à l’aide de modèles hybrides (HMC-MLP), Maillet B., Oltéanu M., Rynkiewicz J., Revue d’économie politique, n.4, p. 489-506 (2004).
Modèles de réseaux de neurones pour l’analyse des séries temporelles ou la régression. J. Rynkiewicz, M. Cottrell, M. Mangeas, J.F. Yao. Revue d’intelligence artificielle, 3-4, Paris, Hermès, p. 317-332, (2001).