June 7, 2024, 4:42 a.m. | Sunwoong Yang, Ricardo Vinuesa, Namwoo Kang

cs.LG updates on arXiv.org arxiv.org

arXiv:2406.03789v1 Announce Type: new
Abstract: This study aims to overcome the conventional deep-learning approaches based on convolutional neural networks, whose applicability to complex geometries and unstructured meshes is limited due to their inherent mesh dependency. We propose novel approaches to improve mesh-agnostic spatio-temporal prediction of transient flow fields using graph U-Nets, enabling accurate prediction on diverse mesh configurations. Key enhancements to the graph U-Net architecture, including the Gaussian mixture model convolutional operator and noise injection approaches, provide increased flexibility in …

abstract arxiv convolutional convolutional neural networks cs.ai cs.lg enabling fields flow graph mesh meshes networks neural networks novel physics.flu-dyn prediction study temporal type unstructured

Senior Data Engineer

@ Displate | Warsaw

Solution Architect

@ Philips | Bothell - B2 - Bothell 22050

Senior Product Development Engineer - Datacenter Products

@ NVIDIA | US, CA, Santa Clara

Systems Engineer - 2nd Shift (Onsite)

@ RTX | PW715: Asheville Site W Asheville Greenfield Site TBD , Asheville, NC, 28803 USA

System Test Engineers (HW & SW)

@ Novanta | Barcelona, Spain

Senior Solutions Architect, Energy

@ NVIDIA | US, TX, Remote