mlsanityData to 7 Oct 2026

Defining error accumulation in ML atmospheric simulators

arXiv 2405.14714 ↗First posted 23 May 2024cs.LGcited by 5

Not on the map. Why?

Abstract (arXiv)

Machine learning (ML) has recently shown significant promise in modelling atmospheric systems, such as the weather. Many of these ML models are autoregressive, and error accumulation in their forecasts is a key problem. However, there is no clear definition of what `error accumulation' actually entails. In this paper, we propose a definition and an associated metric to measure it. Our definition distinguishes between errors which are due to model deficiencies, which we may hope to fix, and those due to the intrinsic properties of atmospheric systems (chaos, unobserved variables), which are not fixable. We illustrate the usefulness of this definition by proposing a simple regularization loss penalty inspired by it. This approach shows performance improvements (according to RMSE and spread/skill) in a selection of atmospheric systems, including the real-world weather prediction task.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 24 of its 49 references that are in the map. First posted on arXiv 23 May 2024.

Builds on

24 of 49 references.

References · 24

  1. 2022Scalable Diffusion Models with Transformerscited by 5,931
  2. 2017Deep reinforcement learning from human preferencescited by 3,846
  3. 2020Learning to summarize from human feedbackcited by 2,562
  4. 2017FiLM: Visual Reasoning with a General Conditioning Layercited by 2,286
  5. 2021Cascaded Diffusion Models for High Fidelity Image Generationcited by 1,155
  6. 2023Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecturecited by 982
  7. 2020Learning to Simulate Complex Physics with Graph Networkscited by 867
  8. 2015Scheduled Sampling for Sequence Prediction with Recurrent Neural Networkscited by 779
  9. 2022FourCastNet: A Global Data-driven High-resolution Weather Model using Adaptive Fourier Neural Operatorscited by 748
  10. 2015Sequence Level Training with Recurrent Neural Networkscited by 568
  11. 2016f-GAN: Training Generative Neural Samplers using Variational Divergence Minimizationcited by 504
  12. 2020WeatherBench: A benchmark dataset for data-driven weather forecastingcited by 272
12 more
  1. 2022GraphCast: Learning skillful medium-range global weather forecastingcited by 261
  2. 2023FuXi: A cascade machine learning forecasting system for 15-day global weather forecastcited by 237
  3. 2016Professor Forcing: A New Algorithm for Training Recurrent Networkscited by 214
  4. 2023FengWu: Pushing the Skillful Global Medium-range Weather Forecast beyond 10 Days Leadcited by 170
  5. 2023GenCast: Diffusion-based ensemble forecasting for medium-range weathercited by 169
  6. 2019Generalization in Generation: A closer look at Exposure Biascited by 123
  7. 2023Scaling transformer neural networks for skillful and reliable medium-range weather forecastingcited by 106
  8. 2019Machine Learning for Stochastic Parameterization: Generative Adversarial Networks in the Lorenz '96 Modelcited by 45
  9. 2020Hierarchical Deep Learning of Multiscale Differential Equation Time-Stepperscited by 37
  10. 2021Combining distribution-based neural networks to predict weather forecast probabilitiescited by 18
  11. 2017SEARNN: Training RNNs with Global-Local Lossescited by 17
  12. 2022Using Probabilistic Machine Learning to Better Model Temporal Patterns in Parameterizations: a case study with the Lorenz 96 modelcited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 1; 2025: 1; 2026: 3.

5 papers.

  1. 2025Thermalizer: Stable autoregressive neural emulation of spatiotemporal chaoscited by 21
  2. 2024CSI-BERT2: A BERT-inspired Framework for Efficient CSI Prediction and Classification in Wireless Communication and Sensingcited by 11
  3. 2026Towards Infinitely Long Neural Simulations: Self-Refining Neural Surrogate Models for Dynamical Systemscited by 1
  4. 2026Adversarial Error Correction for Visual Autoregressive Generationcited by 1
  5. 2026Entropy-Guided k-Guard Sampling for Long-Horizon Autoregressive Video Generationcited by 0