Implementation of finite difference simulation of 2D lid-driven cavity flow in NumPy

Implementation of PINNs in TensorFlow 2

Dynamic normalization with and without bias correction for physics-informed neural networks

Reliable and efficient inverse analysis using physics-informed neural networks with normalized distance functions and adaptive weight tuning

iterative solvers for linear systems

FreeFEM for heat equation.

Implementation of PINNs in PyTorch

Investigation of spectral bias through Fourier transform.

swish can be interpreted as a generalization of linear map and relu.

investigates the relationship between DNN approximation and activation function.