Sampling distributions for generating synthetic data in neural network inverse problem solvers define an implicit regularization operator because the learned operator minimizes empirical risk and conditional expectation minimizes mean-square error.
Jagtap, Zhiping Mao, Nikolaus A
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 2verdicts
UNVERDICTED 2roles
background 1polarities
support 1representative citing papers
A progressive Euler-PINN with geometry-aware loss weighting achieves CFD-comparable pressure and velocity fields for ten NACA6 blades across 30 operating points while cutting computational cost for family-wide screening.
citing papers explorer
-
Sampling Distributions as Regularization in Learned Inverse Problems
Sampling distributions for generating synthetic data in neural network inverse problem solvers define an implicit regularization operator because the learned operator minimizes empirical risk and conditional expectation minimizes mean-square error.