Towards Accelerating Physics Informed Graph Neural Network for Fluid Simulation
SessionPosters Presentation
DescriptionWe introduce a pioneering Multi-GNN Processor Physics-Informed Graph Neural Network (PIGNN) approach which reduced training time of PIGNN to a quarter while maintaining the error rate.
Event Type
Poster
TimeThursday, 5 December 20241:00pm - 2:00pm JST
LocationLobby Gallery (1) & (2), G Block, Level B1
TE
EH
