HyCOP learns policies over compositions of hybrid modules to produce interpretable programs for parametric PDE solution operators with order-of-magnitude OOD gains over monolithic neural operators.
arXiv preprint arXiv:2410.23889 , year=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CE 2verdicts
UNVERDICTED 2representative citing papers
DynaDiff uses weight-graph diffusion with a functional consistency loss and dynamics-informed prompting to generate adapted predictors, reporting 10.78% average accuracy gains over baselines while amortizing adaptation cost offline.
citing papers explorer
-
HyCOP: Hybrid Composition Operators for Interpretable Learning of PDEs
HyCOP learns policies over compositions of hybrid modules to produce interpretable programs for parametric PDE solution operators with order-of-magnitude OOD gains over monolithic neural operators.
-
Generative Adaptation of Dynamics to Environmental Shifts via Weight-space Diffusion
DynaDiff uses weight-graph diffusion with a functional consistency loss and dynamics-informed prompting to generate adapted predictors, reporting 10.78% average accuracy gains over baselines while amortizing adaptation cost offline.