Physics Informed Machine Learning: High-Level Overview of AI and ML in Science and Engineering

Physics Informed Machine Learning: High-Level Overview of AI and ML in Science and Engineering" is a standout introductory resource that bridges the gap between traditional scientific modeling and modern machine learning, ideal for students, researchers, and engineering professionals. The lecture structures ML workflows into five key stages—problem formulation, data curation, architecture design, loss function crafting, and optimization—while demonstrating how physical principles like conservation laws and symmetries can be integrated into each step to build more interpretable, generalizable models that perform well with sparse or noisy data.