April 25: I defended my "Habilitation à Diriger des Recherches" (HDR)
Advances in PAC-Bayesian theory from generalization bounds to supervised and transfer learning algorithms (in french)
May 24: Accepted paper at ECML-PKDD 2024
A Theoretically Grounded Extension of Universal Attacks from the Attacker's Viewpoint
May 24: Accepted paper at ECML-PKDD 2024
A Theoretically Grounded Extension of Universal Attacks from the Attacker's Viewpoint
May 24: A communication accepted paper at CAp 2024
Une borne PAC-Bayésienne sur une mesure de risque pour l'apprentissage équitable
Jan 24: Accepted paper at AISTATS 2024
Leveraging PAC-Bayes Theory and Gibbs Distributions for Generalization Bounds with Complexity Measures
Dec 22: Paul Viallard defended his Ph.D. Congratz ! Beyond PAC-Bayesian Bounds: From Disintegration to Novel Bounds
May 22: 2 communications accepter at CAp'22
I was a meta-reviewer for ECML-PKDD'22
March 22: I was an external examinator of the PhD Thesis of Luxin Zhang --- Well done Model Agnostic Domain Adaptation: Application to Fraud Detection
I was a reviewer for ICML 2022
Two papers accepted at NeurIPS 2021
Learning Stochastic Majority Votes by Minimizing a PAC-Bayes Generalization Bound
A PAC-Bayes Analysis of Adversarial Robustness
I am the publicity chair of ECML-PKDD 2022
I am a membre of CNU 27
the french National Council of Universities for computer science
June 21: A paper accepted at ECML-PKDD'21
Self-Bounding Majority Vote Learning Algorithms by the Direct Minimization of a Tight PAC-Bayesian C-Bound
May 21: 3 communcations accepted at CAp'21
I was a reviewer for ICML 2021
Dec. 20: Léo Gautheron defended his Ph.D. Well Done.
June 20: A paper accepted at ECML-PKDD'20
Landmark-based Ensemble Learning with Random Fourier Features and Gradient Boosting
May 20: Two communications accepted at CAp'20
Apr. 20: A short version of our domain adaptation book is avalaible on ArXiV
Metric Learning from Imbalanced Data with Generalization Guarantees
A paper accepted in Pattern Recognition Letters
Metric Learning from Imbalanced Data with Generalization Guarantees
A communication accepted to ML with Guarantees
workshop @NeurIPS'19
Interpreting Neural Networks as Majority Votes through the PAC-Bayesian Theory