Scaling Transformers for Skillful and Reliable Medium-range Weather Forecasting (Papers Track) Spotlight
Tung Nguyen (University of California, Los Angeles); Rohan Shah (Carnegie Mellon University); Hritik Bansal (UCLA); Troy Arcomano (Argonne National Laboratory); Sandeep Madireddy (Argonne National Laboratory); Romit Maulik (Argonne National Laboratory); Veerabhadra Kotamarthi (Argonne National Laboratory); Ian Foster (Computation Institute); Aditya Grover (UCLA)
Abstract
Weather forecasting is a fundamental problem for anticipating and mitigating the impacts of climate change. Recently, data-driven approaches for weather forecasting based on deep learning have shown great promise, achieving accuracies that are competitive with operational systems. However, those methods often employ complex, customized architectures without sufficient ablation analysis, making it difficult to understand what truly contributes to their success. Here we introduce Stormer, a simple transformer model that achieves state-of-the-art performance on weather forecasting with minimal changes to the standard transformer backbone. We identify the key components of Stormer through careful empirical analyses, including weather-specific embedding, randomized dynamics forecast, and pressure-weighted loss. At the core of Stormer is a randomized forecasting objective that trains the model to forecast the weather dynamics over varying time intervals. During inference, this allows us to produce multiple forecasts for a target lead time and combine them to obtain better forecast accuracy. On WeatherBench 2, Stormer performs competitively at short to medium-range forecasts and outperforms current methods beyond 7 days, while requiring orders-of-magnitude less training data and compute. Additionally, we demonstrate Stormer’s favorable scaling properties, showing consistent improvements in forecast accuracy with increases in model size and training tokens.