Online Reinforcement Learning in the Met Office Unified Model through Distributed Model-Agent Coupling

• URL: http://arxiv.org/abs/2609.02566v1
• Date: Wed, 02 Sep 2026 13:17:10 GMT
• Title: Online Reinforcement Learning in the Met Office Unified Model through Distributed Model-Agent Coupling
• Abstract summary: Machine-learnt corrections can complement numerical weather prediction only if they adapt to the evolving model state.<n>We couple the Met Office (UKMO) Unified Model (UM) with distributed RL agents through rank-local tensors.<n>We show that the learnt policy reduces Z$_500$ MAE in four of six latitude bands, including reductions of 45.8% and 40.8% in the northern and southern tropics.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Machine-learnt corrections can complement numerical weather prediction only if they adapt to the evolving model state while preserving dynamical consistency and numerical stability. To test this within a global forecasting model, we couple the Met Office (UKMO) Unified Model (UM) with distributed RL agents through rank-local tensors. A DDPG actor shares weights across the 70 vertical model levels of each atmospheric column and applies bounded potential-temperature corrections to the model tendencies. Across ten nudged training forecasts, nudging calculations towards the UKMO operational analysis provides an immediate counterfactual target. The frozen policy is then evaluated in a non-nudged forecast for inference. The coupled workflow successfully completes training and remains numerically stable in the evaluated case. Relative to a matched native UM forecast at +6 h, the learnt policy reduces Z$_{500}$ MAE in four of six latitude bands, including reductions of 45.8% and 40.8% in the northern and southern tropics. MSLP error too decreases in three bands, with a maximum reduction of 27.3% at 0-30°N. This single-case experiment demonstrates significant promise and feasibility of distributed online learning followed by non-nudged inference, laying the groundwork for RL-based bias correction and parametrisations within operational systems.

Related papers

• Stochastically Perturbed Weights: Ensembles from Deterministic Machine-Learning Weather Models [0.0]
  Many machine-learning weather models (MLWMs) are deterministic, producing a single forecast with no estimate of its own uncertainty.<n>A growing family of trained-probabilistic models generate calibrated ensembles directly, at the price of a dedicated training run.<n>We ask how much uncertainty can be extracted from a deterministic checkpoint that already exists, without retraining it.
  arXiv  Detail & Related papers  (2026-09-08T08:18:40Z)
• On-Policy Self-Distillation in Diffusion Models [89.32656931795293]
  Reinforcement learning can align diffusion models with human preferences and task-specific objectives, but endpoint rewards do not specify how an intermediate denoising prediction should change.<n>We introduce DiffusionOPSD as an on-policy self-distillation framework that converts image-level reward guidance into explicit targets for clean-output predictions at sampled queries.<n>Across SD 3.5-M and the step-distilled Z-Image-Turbo, our approach achieves the best final held-out scores in 19 of 20 reward-matched settings across two backbones and ten evaluators.
  arXiv  Detail & Related papers  (2026-08-25T14:55:24Z)
• Enhancing the Forecasting Capability of Multi-Model Blending Algorithms for Extreme Precipitation via Joint Use of Station and Gridded Observations [10.058122938830389]
  Traditional precipitation multi-model blending algorithms perform pixel-by-pixel blending on the forecast field based on weights.<n>This study proposes a two-stage framework: probability classification followed by value reconstruction, to blend forecasts from six major NWP models.<n>A novel station-grid joint supervision mechanism is introduced by integrating observations from 2411 national meteorological stations in China into the loss function.
  arXiv  Detail & Related papers  (2026-07-06T09:35:58Z)
• A Trainable-by-Parts Operator Learning Framework: Bridging DeepONet and Karhunen-Loeve Expansions for Large-Scale Applications [46.07145155913717]
  We propose a scalable operator-learning framework based on the Karhunen-Loeve Deep Neural Network (KL-DNN)<n>The model is trained on a dataset comprising 100 samples of large-scale simulations in a three-dimensional domain with 1.7 million cells and 50 time steps.
  arXiv  Detail & Related papers  (2026-06-26T18:15:24Z)
• ARC-STAR: Auditable Post-Hoc Correction for PDE Foundation Models [47.9430487327673]
  ARC-STAR organizes correction into three stages: a global corrector removes broad solver bias, a blockwise local refiner cleans the post-global residual, and, at deployment, a label-free score routes refinement to high-risk blocks under a compute budget.<n>ARC-STAR is the only method that cuts velocity rollout error by at least 36x over raw Poseidon on every cell.
  arXiv  Detail & Related papers  (2026-05-21T09:26:16Z)
• Meteorology-Driven GPT4AP: A Multi-Task Forecasting LLM for Atmospheric Air Pollution in Data-Scarce Settings [4.364999214109123]
  Meteorology-Driven GPT for Air Pollution (GPT4AP) based on a pre-trained GPT-2 backbone and Gaussian rank-stabilized low-rank adaptation (rsLoRA)<n>Model is evaluated on six real-world air quality monitoring datasets under few-shot, zero-shot, and long-term forecasting settings.
  arXiv  Detail & Related papers  (2026-03-31T16:37:06Z)
• Leap+Verify: Regime-Adaptive Speculative Weight Prediction for Accelerating Neural Network Training [0.0]
  We introduce Leap+Verify to accelerate neural network training.<n>It applies speculative execution -- predicting future model weights and validating predictions before acceptance.<n>We evaluate Leap+Verify on GPT-2 124M and Qwen 2.5-1.5B trained on WikiText-103 across five random seeds.
  arXiv  Detail & Related papers  (2026-02-23T08:01:44Z)
• Agentic World Modeling for 6G: Near-Real-Time Generative State-Space Reasoning [70.56067503630486]
  We argue that sixth-generation (6G) intelligence is not fluent token prediction but calibrated the capacity to imagine and choose.<n>We show that WM-MS3M cuts mean absolute error (MAE) by 1.69% versus MS3M with 32% fewer parameters and similar latency, and achieves 35-80% lower root mean squared error (RMSE) than attention/hybrid baselines with 2.3-4.1x faster inference.
  arXiv  Detail & Related papers  (2025-11-04T17:22:22Z)
• Squeeze the Soaked Sponge: Efficient Off-policy Reinforcement Finetuning for Large Language Model [56.92219181993453]
  We propose Reincarnating Mix-policy Proximal Policy Gradient (ReMix) to enable on-policyRFT methods like PPO and GRPO to leverage off-policy data.<n>ReMix consists of three major components: (1) Mix-policy proximal policy gradient with an increased Update-To-Data (UTD) ratio for efficient training; (2) KL-Convex policy constraint to balance the trade-off between stability and flexibility; (3) Policy reincarnation to achieve a seamless transition from efficient early-stage learning to steady improvement.
  arXiv  Detail & Related papers  (2025-07-09T14:29:45Z)
• 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)
• Attention-Based Ensemble Pooling for Time Series Forecasting [55.2480439325792]
  We propose a method for pooling that performs a weighted average over candidate model forecasts. We test this method on two time-series forecasting problems: multi-step forecasting of the dynamics of the non-stationary Lorenz 63 equation, and one-step forecasting of the weekly incident deaths due to COVID-19.
  arXiv  Detail & Related papers  (2023-10-24T22:59:56Z)
• A generalised multi-factor deep learning electricity load forecasting model for wildfire-prone areas [0.0]
  This paper proposes a generalised and robust multi-factor Gated Recurrent Unit (GRU) based Deep Learning (DL) model to forecast electricity load in distribution networks during wildfire seasons. Compared to the regular use of instantaneous temperature, the Mean Absolute Percentage Error (MAPE) is decreased by 30.73%. The proposed method achieves an average forecast MAPE of around 3%, giving a potential annual energy saving of AU$80.46 million for the state of Victoria.
  arXiv  Detail & Related papers  (2023-04-21T00:42:24Z)
• Stabilizing Machine Learning Prediction of Dynamics: Noise and Noise-inspired Regularization [58.720142291102135]
  Recent has shown that machine learning (ML) models can be trained to accurately forecast the dynamics of chaotic dynamical systems. In the absence of mitigating techniques, this technique can result in artificially rapid error growth, leading to inaccurate predictions and/or climate instability. We introduce Linearized Multi-Noise Training (LMNT), a regularization technique that deterministically approximates the effect of many small, independent noise realizations added to the model input during training.
  arXiv  Detail & Related papers  (2022-11-09T23:40:52Z)
• Sub-seasonal forecasting with a large ensemble of deep-learning weather prediction models [6.882042556551611]
  We present an ensemble prediction system using a Deep Learning Weather Prediction (DLWP) model. This model uses convolutional neural networks (CNNs) on a cubed sphere grid to produce global forecasts. Ensemble spread is primarily produced by randomizing the CNN training process to create a set of 32 DLWP models.
  arXiv  Detail & Related papers  (2021-02-09T20:14:43Z)

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.