Papers

See also my Google scholar page.

My research lies in machine learning with focus on…

Hawkes processes

• Nonparametric estimation of Hawkes processes with RKHSs (2024), A. Bonnet, M. Sangnier.

• Spectral analysis for noisy Hawkes processes inference (2024), A. Bonnet, F. Cheysson, M. Martinez Herrera, M. Sangnier. GitHub

• Inference of multivariate exponential Hawkes processes with inhibition and application to neuronal activity (2023), A. Bonnet, M. Martinez Herrera, M. Sangnier. Statistics and Computing.

• Maximum Likelihood Estimation for Hawkes Processes with self-excitation or inhibition (2021), A. Bonnet, M. Martinez Herrera, M. Sangnier. Statistics and Probability Letters. GitHub.

Generative adversarial networks

• Approximating Lipschitz continuous functions with GroupSort neural networks (2021), G. Biau, M. Sangnier, U. Tanielian. International Conference on Artificial Intelligence and Statistics (AISTATS).

• Some theoretical insights into Wasserstein GANs (2021), G. Biau, M. Sangnier, U. Tanielian. Journal of Machine Learning Research.

• Some Theoretical Properties of GANs (2020), G. Biau, B. Cadre, M. Sangnier, U. Tanielian. The Annals of Statistics.

Supervised learning

• Proximal boosting: aggregating weak learners to minimize non-differentiable losses (2023), E. Fouillen, C. Boyer, M. Sangnier. Neurocomputing. GitHub.

• Infinite-task learning with vector-valued RKHSs (2018), R. Brault, A. Lambert, Z. Szabo, M. Sangnier, F. d'Alché-Buc. International Conference on Artificial Intelligence and Statistics (AISTATS).

• Output Fisher Embedding Regression (2018), M. Djerrab, A. Garcia, M. Sangnier, F. d'Alché-Buc. Machine Learning.

• Data sparse nonparametric regression with \(\epsilon\)-insensitive losses (2017), M. Sangnier, O. Fercoq, F. d'Alché-Buc. Asian Conference on Machine Learning (ACML). GitHub.

• Joint quantile regression in vector-valued RKHSs (2016), M. Sangnier, O. Fercoq, F. d'Alché-Buc. Neural Information Processing Systems (NIPS). GitHub.

Fluid flows

• Reduced Basis’ Acquisition by a Learning Process for Rapid On-line Approximation of Solution to PDE's: Laminar Flow Past a Backstep (2017), P. Gallinari, Y. Maday, M. Sangnier, O. Schwander, T. Taddei. Archives of Computational Methods in Engineering.

Signal recognition

• Early and reliable event detection using proximity space representation (2016), M. Sangnier, J. Gauthier, A. Rakotomamonjy. International Conference on Machine Learning (ICML). GitHub.

• Filter bank learning for signal classification (2015), M. Sangnier, J. Gauthier, A. Rakotomamonjy. Signal Processing.

• Early frame-based detection of acoustic scenes (2015), M. Sangnier, J. Gauthier, A. Rakotomamonjy. International Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA). GitHub.

• Kernel learning as minimization of the single validation estimate (2014), M. Sangnier, J. Gauthier, A. Rakotomamonjy. International Workshop on Machine Learning for Signal Processing (MLSP).

• Filter Bank Kernel Learning for nonstationary signal classification (2013), M. Sangnier, J. Gauthier, A. Rakotomamonjy. 38th International Conference on Acoustics, Speech, and Signal Processing (ICASSP).

• Learning a time-frequency representation adapted to signal classification through positive semidefinite optimization (2013), M. Sangnier, J. Gauthier, A. Rakotomamonjy. XXIVème Colloque GRETSI.

• Comparison of features for the classification of sparse shift-invariant decompositions (2013), Q. Barthélemy, M. Sangnier, A. Larue, J. Mars. XXIVème Colloque GRETSI.

Polarimetric images

• Statistics on High Resolution urban polarimetric images: Application to segmentation and classification (2012), N. Trouvé, M. Sangnier, E. Colin Koeniguer. 9th European Conference on Synthetic Aperture Radar (EUSAR).