PCELM overcomes nonconvex optimization difficulties in neural approximations of Stefan problems by using an initial rough fit followed by a perturbation-based convex correction step that improves relative L2 accuracy by 2-6 orders of magnitude.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.NA 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
PCELM: Perturbation-Correction Extreme Learning Machine for the Stefan problem
PCELM overcomes nonconvex optimization difficulties in neural approximations of Stefan problems by using an initial rough fit followed by a perturbation-based convex correction step that improves relative L2 accuracy by 2-6 orders of magnitude.