1. Mathias Valla. Machine Learning,6e Saga de l’actuariat du groupe APRIL, Fev 2018, Lyon, France. [Slides]
2. Mathias Valla. Towards tree-based time-to-event models with longitudinal data, Literature review and applications to churn analysis, Chaire DIALog’s technical seminar, Nov 2021, Paris, France.
3. Mathias Valla. Longitudinal trees: an overview of tree-based models for longitudinal analysis with time-varying covariates2022 EDEG Workshop, Apr 2022, Lyon, France.
4. Mathias Valla. Including Customer Lifetime Value in tree-based lapse management strategy. MLISTRAL, Sep 2022, Marseille, France. [Slides]
   
5. Mathias Valla. Tree-based models, longitudinal covariates and customer behaviour predictions, Chaire DIALog’s technical seminar, Nov 2022, Paris, France.
   
6. Mathias Valla and Pierrick Piette. Améliorer un modèle de Machine Learning par la personnalisation : fonction de perte et enrichissement par les données.100% Actuaires 100% Data Science, Institut des actuaires, Nov 2022, Paris, France. [Slides]
   
7. Mathias Valla. Ma thèse en 180 secondes (3 minutes thesis), MT180s regional finals, Mar 2023, Lyon, France.
   -> Jury’s second pize [Vidéo]
   
8. Mathias Valla. Longitudinal analysis: contributions in life insurance and applications, Chaire DIALog’s technical seminar, Nov 2023, Paris, France. [Vidéo]
   
9. Mathias Valla. Temporal dynamics in tree-based models and applications to lapse behaviour in life insurance,Utilisation de la data pour la modélisation et la transformation des comportements en assurance, Galea and associates Data Lab, Feb 2024, Paris, France
   
10. Mathias Valla. Presentation of EAJ Issues 14/1-2, Including Customer Lifetime Value in tree-based lapse management strategy, hosted by the European Actuarial Journal, Oct 2024, Online. [Slides]
11. Mathias Valla, Leonie Le Bastard, Jose Garrido. Time-penalised tree : un algorithme temporel interprétable appliqué aux risques climatiques. Journée 100% Actuaires 100% Data sciences 100% Durabilité, Institut des actuaires, Nov 2024, Paris, France.
12. Mathias Valla and José Garrido. Feature and quantile selection for the actuarial climate index: Everything, everywhere, all at once (1/2). Insurance Data Science Conference, Bayes Business School, Jun 2025, London, United Kingdom. [Slides]
13. Mathias Valla and José Garrido. Feature and quantile selection for the actuarial climate index: Everything, everywhere, all at once (2/2). Climate Change and Insurance 2025, The National Robotarium, Heriot-Watt University Edinburgh Campus, Scotland, United Kingdom