00:00:00 / 00:00:00

Coupling rare event algorithms and deep neural network to predict extreme heat waves

De Freddy Bouchet

Apparaît dans la collection : On Future Synergies for Stochastic and Learning Algorithms / Sur les synergies futures autour des algorithmes d'apprentissage et stochastiques

Extreme events are of primarily importance for understanding the impact of climate change. However, because they are too rare and realistic models are too complex, traditional deep neural networks are inefficient for predictions. We cope with this lack of data using rare event simulations. From the best climate models, we oversample extremely rare events and obtain several hundreds more events than with usual climate runs, at a fixed numerical cost. Coupled with deep neural networks this approach improves drastically the prediction of extreme heat waves.

Informations sur la vidéo

Données de citation

  • DOI 10.24350/CIRM.V.19817403
  • Citer cette vidéo Bouchet, Freddy (27/09/2021). Coupling rare event algorithms and deep neural network to predict extreme heat waves. CIRM. Audiovisual resource. DOI: 10.24350/CIRM.V.19817403
  • URL https://dx.doi.org/10.24350/CIRM.V.19817403

Bibliographie

  • RAGONE, Francesco, WOUTERS, Jeroen, et BOUCHET, Freddy. Computation of extreme heat waves in climate models using a large deviation algorithm. Proceedings of the National Academy of Sciences, 2018, vol. 115, no 1, p. 24-29. - https://doi.org/10.1073/pnas.1712645115
  • LUCENTE, Dario, ROLLAND, Joran, HERBERT, Corentin, et al. Coupling rare event algorithms with data-based learned committor functions using the analogue Markov chain. arXiv preprint arXiv:2110.05050, 2021. - https://arxiv.org/abs/2110.05050
  • JACQUES-DUMAS, Valérian, RAGONE, Francesco, BORGNAT, Pierre, et al. Deep Learning-based Extreme Heatwave Forecast. arXiv preprint arXiv:2103.09743, 2021. - https://arxiv.org/abs/2103.09743

Dernières questions liées sur MathOverflow

Pour poser une question, votre compte Carmin.tv doit être connecté à mathoverflow

Poser une question sur MathOverflow




Inscrivez-vous

  • Mettez des vidéos en favori
  • Ajoutez des vidéos à regarder plus tard &
    conservez votre historique de consultation
  • Commentez avec la communauté
    scientifique
  • Recevez des notifications de mise à jour
    de vos sujets favoris
Donner son avis