Fixing the Double Penalty in Data-Driven Weather Forecasting Through a Modified Spherical Harmonic Loss Function
- URL: http://arxiv.org/abs/2501.19374v1
- Date: Fri, 31 Jan 2025 18:23:45 GMT
- Title: Fixing the Double Penalty in Data-Driven Weather Forecasting Through a Modified Spherical Harmonic Loss Function
- Authors: Christopher Subich, Syed Zahid Husain, Leo Separovic, Jing Yang,
- Abstract summary: Fine-tuning the GraphCast model results in sharp deterministic weather forecasts, an increase of the model's effective resolution from 1,250km to 160km, and improvements to ensemble spread.
- Score: 2.4020585213586387
- License:
- Abstract: Recent advancements in data-driven weather forecasting models have delivered deterministic models that outperform the leading operational forecast systems based on traditional, physics-based models. However, these data-driven models are typically trained with a mean squared error loss function, which causes smoothing of fine scales through a "double penalty" effect. We develop a simple, parameter-free modification to this loss function that avoids this problem by separating the loss attributable to decorrelation from the loss attributable to spectral amplitude errors. Fine-tuning the GraphCast model with this new loss function results in sharp deterministic weather forecasts, an increase of the model's effective resolution from 1,250km to 160km, improvements to ensemble spread, and improvements to predictions of tropical cyclone strength and surface wind extremes.
Related papers
- Physics-guided Active Sample Reweighting for Urban Flow Prediction [75.24539704456791]
Urban flow prediction is a nuanced-temporal modeling that estimates the throughput of transportation services like buses, taxis and ride-driven models.
Some recent prediction solutions bring remedies with the notion of physics-guided machine learning (PGML)
We develop a atized physics-guided network (PN), and propose a data-aware framework Physics-guided Active Sample Reweighting (P-GASR)
arXiv Detail & Related papers (2024-07-18T15:44:23Z) - CaFA: Global Weather Forecasting with Factorized Attention on Sphere [7.687215328455751]
We propose a factorized-attention-based model tailored for spherical geometries to mitigate this issue.
The deterministic forecasting accuracy of the proposed model on $1.5circ$ and 0-7 days' lead time is on par with state-of-the-art purely data-driven machine learning weather prediction models.
arXiv Detail & Related papers (2024-05-12T23:18:14Z) - FuXi-ENS: A machine learning model for medium-range ensemble weather forecasting [16.562512279873577]
We introduce FuXi-ENS, an advanced ML model designed to deliver 6-hourly global ensemble weather forecasts up to 15 days.
FuXi-ENS runs at a significantly increased spatial resolution of 0.25textdegree, incorporating 5 atmospheric variables at 13 pressure levels, along with 13 surface variables.
Results demonstrate that FuXi-ENS outperforms ensemble forecasts from the ECMWF, a world leading NWP model, in the CRPS of 98.1% of 360 variable and forecast lead time combinations.
arXiv Detail & Related papers (2024-05-09T17:15:09Z) - ExtremeCast: Boosting Extreme Value Prediction for Global Weather Forecast [57.6987191099507]
We introduce Exloss, a novel loss function that performs asymmetric optimization and highlights extreme values to obtain accurate extreme weather forecast.
We also introduce ExBooster, which captures the uncertainty in prediction outcomes by employing multiple random samples.
Our solution can achieve state-of-the-art performance in extreme weather prediction, while maintaining the overall forecast accuracy comparable to the top medium-range forecast models.
arXiv Detail & Related papers (2024-02-02T10:34:13Z) - Learning Robust Precipitation Forecaster by Temporal Frame Interpolation [65.5045412005064]
We develop a robust precipitation forecasting model that demonstrates resilience against spatial-temporal discrepancies.
Our approach has led to significant improvements in forecasting precision, culminating in our model securing textit1st place in the transfer learning leaderboard of the textitWeather4cast'23 competition.
arXiv Detail & Related papers (2023-11-30T08:22:08Z) - Residual Corrective Diffusion Modeling for Km-scale Atmospheric Downscaling [58.456404022536425]
State of the art for physical hazard prediction from weather and climate requires expensive km-scale numerical simulations driven by coarser resolution global inputs.
Here, a generative diffusion architecture is explored for downscaling such global inputs to km-scale, as a cost-effective machine learning alternative.
The model is trained to predict 2km data from a regional weather model over Taiwan, conditioned on a 25km global reanalysis.
arXiv Detail & Related papers (2023-09-24T19:57:22Z) - Improved Loss Function-Based Prediction Method of Extreme Temperatures
in Greenhouses [3.4893854610476267]
The prediction of extreme greenhouse temperatures to which crops are susceptible is essential in the field of greenhouse planting.
Due to the lack of extreme temperature data in datasets, it is challenging for models to accurately predict it.
We propose an improved loss function, which is suitable for a variety of machine learning models.
arXiv Detail & Related papers (2021-11-02T04:33:15Z) - Data-Based Models for Hurricane Evolution Prediction: A Deep Learning
Approach [0.0]
Many-to-Many RNN storm trajectory prediction models presented here are significantly faster than ensemble models used by the NHC.
A detailed analysis of model forecasting error shows that Many-To-One prediction models are less accurate than Many-To-Many models owing to compounded error accumulation.
arXiv Detail & Related papers (2021-10-30T00:31:48Z) - Back2Future: Leveraging Backfill Dynamics for Improving Real-time
Predictions in Future [73.03458424369657]
In real-time forecasting in public health, data collection is a non-trivial and demanding task.
'Backfill' phenomenon and its effect on model performance has been barely studied in the prior literature.
We formulate a novel problem and neural framework Back2Future that aims to refine a given model's predictions in real-time.
arXiv Detail & Related papers (2021-06-08T14:48:20Z) - Churn Reduction via Distillation [54.5952282395487]
We show an equivalence between training with distillation using the base model as the teacher and training with an explicit constraint on the predictive churn.
We then show that distillation performs strongly for low churn training against a number of recent baselines.
arXiv Detail & Related papers (2021-06-04T18:03:31Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.