Machine learning (ML) is transforming the way computational models are developed and deployed in aeronautical engineering, enabling quicker and more flexible workflows. However, challenges remain in terms of generalization, accuracy, and model validation, especially for industrial adoption. This minisymposium aims to gather recent developments in the use of ML models in aerospace applications. Topics of interest include: • Surrogate models for fast predictions, e.g., for surface aerodynamic quantities; • Data-driven Physics-Informed Neural Networks (DD-PINNs) for solving fluid dynamics problems with embedded physical constraints; • Deep Reinforcement Learning (DRL) for efficient aerodynamic and structural shape optimization [1]; • Validation and verification tools and best practices to assess ML models beyond traditional error metrics [2] The symposium welcomes contributions showcasing methodological advances, benchmark studies, and industrial applications. REFERENCES [1] Ramos, David, et al. "Aerodynamic and structural airfoil shape optimisation via Transfer Learning-enhanced Deep Reinforcement Learning." arXiv preprint arXiv:2505.02634(2025). [2] Lacasa, Lucas, et al. "Towards certification: A complete statistical validation pipeline for supervised learning in industry." Expert Systems with Applications 277 (2025): 127169.