[2505.12167] FABLE: A Localized, Targeted Adversarial Attack on Weather Forecasting Models

Computer Science > Machine Learning arXiv:2505.12167 (cs) [Submitted on 17 May 2025 (v1), last revised 3 Apr 2026 (this version, v2)] Title:FABLE: A Localized, Targeted Adversarial Attack on Weather Forecasting Models Authors:Yue Deng, Asadullah Hill Galib, Xin Lan, Jack Gunn, Pang-Ning Tan, Lifeng Luo View a PDF of the paper titled FABLE: A Localized, Targeted Adversarial Attack on Weather Forecasting Models, by Yue Deng and Asadullah Hill Galib and Xin Lan and Jack Gunn and Pang-Ning Tan and Lifeng Luo View PDF HTML (experimental) Abstract:Deep learning-based weather forecasting (DLWF) models have recently demonstrated significant performance gains over gold-standard physics-based simulation tools. However, these models are potentially vulnerable to adversarial attacks, which raises concerns about their trustworthiness. In this paper, we investigate the feasibility and challenges of applying existing adversarial attack methods to DLWF models and propose a novel framework called FABLE (Forecast Alteration By Localized targeted advErsarial attack) to address them. FABLE performs a 3D discrete wavelet decomposition to disentangle the spatial and temporal components of the data. By regulating the magnitude of adversarial perturbations across different components, FABLE produces adversarial inputs that remain closely aligned with the original inputs while steering the DLWF models toward generating the targeted forecast outcomes. Experimental results on real-world weather data...