Search Results for author: Sashikumaar Ganesan

Found 2 papers, 1 papers with code

FastVPINNs: Tensor-Driven Acceleration of VPINNs for Complex Geometries

no code implementations18 Apr 2024 Thivin Anandh, Divij Ghose, Himanshu Jain, Sashikumaar Ganesan

Variational Physics-Informed Neural Networks (VPINNs) utilize a variational loss function to solve partial differential equations, mirroring Finite Element Analysis techniques.

SParSH-AMG: A library for hybrid CPU-GPU algebraic multigrid and preconditioned iterative methods

2 code implementations30 Jun 2020 Sashikumaar Ganesan, Manan Shah

Further, the performance of CPU-GPU algorithms are compared with the GPU-only implementations to illustrate the significantly lower memory requirements.

Mathematical Software 65F10, 65F50, 65N55, 65Y05

Cannot find the paper you are looking for? You can Submit a new open access paper.