When
2 p.m., Nov. 16, 2022
Interpreting Interpretable Machine Learning: Insights gleaned modeling the velocity gradient tensor in turbulence
Recent application of physics-informed machine learning has found considerable success in modeling the velocity gradient tensor in isotropic turbulence. In this work, we use statistics of the velocity gradient tensor and output of our deep neural network, to find a (physically meaningful!) latent space whilst predicting the non-local pressure Hessian. Come for the turbulence, stay for the statistics!
Math, 402 and Zoom: Link https://arizona.zoom.us/j/83541348598 Password: BB2022