Scientific Machine Learning through the Lens of Physics-Informed Neural Networks

Explore the emerging field of physics-informed machine learning (PIML) in this hour-long webinar focusing on physics-informed neural networks (PINNs). Delve into the capabilities and limitations of PINNs, examining their effectiveness in solving complex scientific problems compared to traditional approaches. Learn about scalable extensions like conservative PINNs (cPINNs) and extended PINNs (XPINNs) for handling big data and large models. Discover a unified framework for causal sweeping strategies and temporal decompositions in PINNs. Gain insights into how PIML addresses challenges in scientific computation, including high-dimensional problems, parameterized PDEs, and efficient inverse problem solving with noisy data incorporation.