Deep Fluids: A Generative Network for Parameterized FluidSimulations

Apr 10, 2020 - 3:00 pm to 4:00 pm
Speaker
Al Zamora (KIPAC) via zoom

Join us this Friday, April 10th at 3pm exclusively on zoom for the next meeting of the Stats and ML Journal Club. This week, Al Zamora will be leading a discussion on CNNs for Fluid sims. See you all there! 

Sean

Title: Deep Fluids: A Generative Network for Parameterized FluidSimulations

Abstract: Over the course of the last few years, machine learning has found its way into many scientific applications. In this presentation, we discuss several applications of deep learning algorithms for simulating fluid flow. We focus on two papers: one which discusses simulating particular parameterized fluid problems, and another which simulates fluid flow around some arbitrarily chosen geometric boundaries (flow around objects). We discuss the errors in these simulations and how they compare with some state-of-the-art solvers as well as other limitations.