Please note that this session was withdrawn and is no longer available in the respective programme. This withdrawal might have been the result of a merge with another session.

ESSI1.2 | Explainable Artificial Intelligence (XAI) for Geosciences
Explainable Artificial Intelligence (XAI) for Geosciences
Convener: Lorenzo NavaECSECS | Co-conveners: Miguel-Ángel Fernández-TorresECSECS, Monique Kuglitsch, Mira KenzhebayECSECS, Arthur Hrast Essenfelder

As Artificial Intelligence (AI) applications in geosciences grow, the quest for understanding Machine Learning (ML) and Deep Learning (DL) models becomes pivotal. This session highlights the critical role of Explainable Artificial Intelligence (XAI) in strengthening our ability to trust, comprehend, and improve AI models. To achieve this, it brings together specialists in geoscience, data science, and AI.
We strongly encourage submissions that employ methods enabling AI systems to furnish lucid and understandable explanations for their decisions.

This multidisciplinary session encompasses contributions related to the following lines of research:
Exploration of novel XAI techniques and methodologies that enhance the transparency and interpretability of ML/DL models used in geosciences.
Real-world case studies where XAI has made substantial progress in understanding and managing specific geoscience tasks and physical processes.
Process understanding via XAI and hybrid, physically-informed modeling.
Quantitative evaluation and comparison of the effectiveness of XAI models.
Strategies towards more scientifically valuable explanations (e.g., use of Large Language Models (LLMs) for XAI).