What changed
Applies AFNO to autoregressive high-resolution global weather fields and develops feature/data-parallel GPU training with optimized spectral operations, I/O and launch overhead.
Scientific scope
Applies AFNO to autoregressive high-resolution global weather fields and develops feature/data-parallel GPU training with optimized spectral operations, I/O and launch overhead. Two 2022 sources bundled with separate 13/9-person bylines; SC account shares contribution among Subramanian/Harrington/Pathak, not automatically first author. Forecast speed is inference-only on specific hardware/variables; ERA5 reanalysis/training cost and fewer vertical variables matter. Distributed timings include exposed I/O/validation but exclude setup/warm-up/capture and extrapolate some 80-epoch costs from five epochs. No general climate or exact physical-simulation certificate.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Two 2022 sources bundled with separate 13/9-person bylines; SC account shares contribution among Subramanian/Harrington/Pathak, not automatically first author. Forecast speed is inference-only on specific hardware/variables; ERA5 reanalysis/training cost and fewer vertical variables matter. Distributed timings include exposed I/O/validation but exclude setup/warm-up/capture and extrapolate some 80-epoch costs from five epochs. No general climate or exact physical-simulation certificate.
Contributors
- Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.
- FourCastNet: A Global Data-driven High-resolution Weather Model using Adaptive Fourier Neural Operators
- FourCastNet: Accelerating Global High-Resolution Weather Forecasting using Adaptive Fourier Neural Operators
Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.
Full scientific record · Browse contribution grades · Request a correction
A