Higher-Order LaSDI: Reduced Order Modeling with Multiple Time Derivatives
By: Robert Stephany, William Michael Anderson, Youngsoo Choi
Solving complex partial differential equations is vital in the physical sciences, but often requires computationally expensive numerical methods. Reduced-order models (ROMs) address this by exploiting dimensionality reduction to create fast approximations. While modern ROMs can solve parameterized families of PDEs, their predictive power degrades over long time horizons. We address this by (1) introducing a flexible, high-order, yet inexpensive finite-difference scheme and (2) proposing a Rollout loss that trains ROMs to make accurate predictions over arbitrary time horizons. We demonstrate our approach on the 2D Burgers equation.
Similar Papers
Rollout-LaSDI: Enhancing the long-term accuracy of Latent Space Dynamics
Machine Learning (CS)
Makes computer predictions of science problems last longer.
RONOM: Reduced-Order Neural Operator Modeling
Machine Learning (CS)
Makes computer models of physics faster and more accurate.
Efficient data-driven regression for reduced-order modeling of spatial pattern formation
Pattern Formation and Solitons
Makes complex patterns appear faster on computers.