Ph 22 - Computational Physics Laboratory III

An introduction to machine learning techniques for modeling and analyzing physical systems. Topics include classification, regression, dimensionality reduction, and the construction of simple neural networks. Students will apply these methods to datasets from physics experiments and simulations, using machine learning to identify patterns, build predictive models, and work with high-dimensional data. The course also includes one reinforcement learning problem, focused on optimizing control in a basic physical system. Emphasis is placed on understanding when and how to apply ML techniques in physics, with attention to interpretability, overfitting, and the use of physical constraints in guiding learning-based approaches.