GrINd: Grid Interpolation Network for Scattered Observations

arXiv:2403.19570v1 Announce Type: new
Abstract: Predicting the evolution of spatiotemporal physical systems from sparse and scattered observational data poses a significant challenge in various scientific domains. Traditional methods rely on dense grid-structured data, limiting their applicability in scenarios with sparse observations. To address this challenge, we introduce GrINd (Grid Interpolation Network for Scattered Observations), a novel network architecture that leverages the high-performance of grid-based models by mapping scattered observations onto a high-resolution grid using a Fourier Interpolation Layer. In the …

abstract arxiv challenge cs.lg data domains evolution grid network scientific structured data systems type