#machine learning
Jun 2026
Starter-Iterator Neural Operator: A Unified Architecture for High-Fidelity Forward and Inverse PDE Problems
This work proposes the Starter-Iterator Neural Operator (SINO), which incorporates the initialization and residual-correction structures of classical iterative solvers into neural operator learning and achieves competitive accuracy and stable performance across the benchmarks considered in this work.
Kui-Lin Qin, Lian-Fang Wang, Xue-Miao Sun et al.
· Journal of Computational Phy... · 1 citation