Reduced Model Formation for 2D Vortex Interactions Using Machine Learning

Haym Hirsh, Tom Ellman, Arunava Banerjee, David Dritschel, Hongbing Mao, and Norm Zabusky

Our work is aimed at developing tools for automatic or semi-automatic formulation of new reduced models of 2D vortex interactions. In particular, we are aiming for models that strike a balance between the goals of high accuracy and low computational expense. In our context, we measure accuracy according to how well the reduced model predicts the global properties of the interacting vortices, such as the total number and mean size of vortices at each point in time, rather than the detailed time-dependent properties of each individual vortex (Carnevale et al. 1992).


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.