PCA-RaNN recasts latent neural operator learning as PCA-reduced random-feature linear regression, achieving 1-3 orders faster training than standard methods on PDE benchmarks while adding conformal uncertainty quantification.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
LNF-NO improves PDE operator learning by multiplicatively fusing a linear component with a nonlinear component, yielding faster training and comparable accuracy across multiple benchmarks.
citing papers explorer
-
Randomized neural operator for parametric PDEs with fast training and conformal uncertainty quantification
PCA-RaNN recasts latent neural operator learning as PCA-reduced random-feature linear regression, achieving 1-3 orders faster training than standard methods on PDE benchmarks while adding conformal uncertainty quantification.
-
Linear-Nonlinear Fusion Neural Operator for Partial Differential Equations
LNF-NO improves PDE operator learning by multiplicatively fusing a linear component with a nonlinear component, yielding faster training and comparable accuracy across multiple benchmarks.