The increasing volume, diversity, and complexity of volcanological datasets are creating new opportunities for Machine Learning and computational approaches to extract information, identify patterns, and characterize changes in volcanic systems. This session welcomes contributions aimed at identifying transitions in volcanic state, characterizing unrest and eruptive processes, and developing automated and near-real-time monitoring and forecasting systems, as well as studies addressing the evaluation, comparison, interpretability, robustness, and transferability of ML models across different datasets and volcanic systems. We particularly encourage contributions from volcanologists, data scientists, and monitoring communities that bridge innovative computational methodologies with sustainable data practices, promoting reproducible and reusable ML workflows and data products. Datasets, derived products, workflows, and computational outputs are encouraged to adhere to FAIR (Findable, Accessible, Interoperable and Reusable) and open data principles. By sharing these through dedicated research infrastructures and services, such as EPOS, their accessibility, interoperability, and reuse are enhanced for the broader volcanological community.