Tag
The paper introduces equation recast, a method that reformulates parametric operator learning as a single canonical operator, enabling zero-shot prediction across new parameter regimes by analytically deriving parameter-induced variations.
This paper compares Fourier spectral differentiation and spatial automatic differentiation in periodic physics-informed neural networks, finding that Fourier methods achieve significant training speedups and memory reductions without compromising accuracy.
This study investigates the benefits of frequency decomposition for Physics-Informed Neural Networks (PINNs) by proposing a dual-branch, spectrally-gated architecture (DBSG-PINN). Ablation experiments on 1D PDE benchmarks indicate that frequency decomposition is most effective on spectrally complex problems, reducing error by up to 59.2%.
RECAST is a machine-learning framework that combines learned correction and super-resolution to restore accuracy in coarse-grid PDE solvers, reducing error by 50-92% across six 1D PDE systems and enabling coarser simulations without losing fidelity.
This paper introduces feature interaction modules based on factorization machines into physics-informed neural networks and neural operators (FM-PINN, FM-Operator, FM-DeepONet) to better capture spatio-temporal variable couplings for solving parameterized PDEs, showing accuracy gains particularly on shock-dominated equations.
Introduces Energy Manifold Natural Gradient Descent (EMNGD), a manifold optimization framework for neural PDE solvers that aligns parameter updates with function-space energy curvature while respecting parameter constraints. Theoretical guarantees and empirical results show improved accuracy and convergence.
Anima Anandkumar highlights that neural operators, despite simple benchmarks, have achieved massive speedups (10,000–million times) in hard real-world problems like high-resolution AI weather modeling (FourCastNet) and nuclear fusion turbulence, referencing a new paper showing learned solvers become more cost-effective as PDE tasks get harder.
This paper establishes quantitative Sobolev approximation bounds for neural operators, proving that operators can be uniformly approximated with explicit complexity-error relations. It validates these theoretical bounds using Fourier Neural Operators on the Burgers' equation, demonstrating that Sobolev-space approximation theory accurately predicts scaling behavior.