Machine learning offers new ways to extract insight from complex, high-dimensional data generated by simulations and experiments. In our lab, we use machine learning as a tool to enhance and complement physical understanding. We focus on identifying patterns, relationships, and reduced representations that improve prediction and interpretability. These methods help us navigate large design spaces and identify key variables that control behavior. By integrating machine learning with physics-based models, we accelerate discovery while maintaining scientific insight. This approach equips our lab members with broadly applicable computational skills.
