3 papers
cs.MS2026
Implementation of tangent linear and adjoint models for neural networks based on a compiler library tool
Sa Xiao, Hao Jing, Honglu Sun +1
This paper presents TorchNWP, a compilation library tool for the efficient coupling of artificial intelligence components and traditional numerical models. It aims to address the i…
cs.LG2026
China Regional 3km Downscaling Based on Residual Corrective Diffusion Model
Honglu Sun, Hao Jing, Zhixiang Dai +4
A fundamental challenge in numerical weather prediction is to efficiently produce high-resolution forecasts. A common solution is applying downscaling methods, which include dynami…
cs.LG2026
Machine learning based radiative parameterization scheme and its performance in operational reforecast experiments
Hao Jing, Sa Xiao, Haoyu Li +2
Radiation is typically the most time-consuming physical process in numerical models. One solution is to use machine learning methods to simulate the radiation process to improve co…