Communication Dans Un Congrès Année : 2025

Rockfall Forecasting using Ensemble Deep Learning and Temporal Gradient-Based Explanations

Résumé

Rockfalls pose a significant risk for infrastructures and human activities. Accurate forecasting is therefore crucial for prevention. While current methods rely on expert hypotheses, AI models, particularly explainable ones, offer a promising way to improve the accuracy of forecasts and provide transparent, trustworthy results. In this paper, an ensemble of multisource neural networks based on InceptionTime modules whose forecasts are explained using a temporal Grad-CAM approach is being developed for the Mont Saint-Eynard cliff, showcasing the potential of AI in this area.

Fichier principal
Vignette du fichier
main.pdf (5.76 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-05311348 , version 1 (13-10-2025)

Licence

Identifiants

  • HAL Id : hal-05311348 , version 1

Citer

Sabrine Bouaziz, Christophe Lin-Kwong-Chon, David Amitrano, Nicolas Méger. Rockfall Forecasting using Ensemble Deep Learning and Temporal Gradient-Based Explanations. International Conference on Data Mining (ICDM), Workshop of Artificial Intelligence for Time Series Analysis (AI4TS): Theory, Algorithms, and Applications, IEEE, Nov 2025, Washington, D.C., United States. ⟨hal-05311348⟩
483 Consultations
308 Téléchargements

Partager

  • More