Select a category to explore research frontiers
Loading categories...
Development of neural network architectures that integrate domain-specific physical laws and differential equations as inductive biases for solving inverse problems and discovering governing equations from observational data.