paper-with-me

홈 › Papers

Adversarial Attacks on Downstream Weather Forecasting Models: Application to Tropical Cyclone Trajectory Prediction

2025-10-11 · Yue Deng, Francisco Santos, Pang-Ning Tan, Lifeng Luo arxiv

Deep learning-based weather forecasting (DLWF) models leverage past weather observations to generate future forecasts, supporting a wide range of downstream applications, including tropical cyclone (TC) prediction. In this paper, we investigate their vulnerability to adversarial attacks, where subtle perturbations to the upstream forecasts can alter the downstream TC trajectory predictions. Although research into adversarial attacks on DLWF models has grown recently, it remains challenging to craft perturbed upstream forecasts that steer the downstream outputs toward attacker-specified trajectories. First, conventional TC detection systems are opaque, non-differentiable black boxes, making standard gradient-based attacks infeasible. Second, the extreme rarity of TC events leads to severe class imbalance problem, making it difficult to develop attack methods for perturbing upstream forecasts that produce realistic-looking cyclone paths aligned with attacker's target trajectories. To overcome these limitations, we propose Cyc-Attack, a novel method for perturbing the upstream forecasts of DLWF models to generate adversarial trajectories. The proposed method uses a differentiable surrogate model to approximate the TC detector's output, enabling the application of gradient-based attacks. Cyc-Attack also employs a skewness-aware loss function with kernel dilation strategy to address the imbalance problem. Finally, a distance-based gradient weighting scheme and regularization are used to constrain the perturbations and eliminate unrealistic-looking trajectories, thereby making the adversarial upstream forecasts less easily detectable. Our experiments show that Cyc-Attack achieves a higher true positive rate in matching the attacker's target trajectories, along with lower false alarm rates and stealthier perturbations than conventional attack methods.

📄 PDF Abstract BibTeX arXiv:2510.10140

Code (0)

등록된 구현이 없습니다.

Tasks

Trajectory PredictionWeather Forecasting

Similar Papers 제목 키워드 기반

Forecasting Fails: Unveiling Evasion Attacks in Weather Prediction Models

2025-12-09 · Huzaifa Arif, Pin-Yu Chen, Alex Gittens, James Diffenderfer 외 arxiv

With the increasing reliance on AI models for weather forecasting, it is imperative to evaluate their vulnerability to adversarial perturbations. This work introduces Weather Adaptive Adversarial Perturbation Optimizatio…

Weather Forecasting

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

2025-05-17 · Yue Deng, Asadullah Hill Galib, Xin Lan, Pang-Ning Tan 외

Deep learning-based weather forecasting models have recently demonstrated significant performance improvements over gold-standard physics-based simulation tools. However, these models are vulnerable to adversarial attack…

Adversarial AttackWeather Forecasting

Guided Diffusion Sampling for Precipitation Forecast Interventions

2026-05-14 · Ayumu Ueyama, Kazuhiko Kawamoto, Hiroshi Kera arxiv

Extreme precipitation causes severe societal and economic damage, and weather control has long been discussed as a potential mitigation strategy. However, to the best of our knowledge, perturbation-based interventions fo…

Weather Forecasting

fakeWeather: Adversarial Attacks for Deep Neural Networks Emulating Weather Conditions on the Camera Lens of Autonomous Systems

2022-05-27 · Alberto Marchisio, Giovanni Caramia, Maurizio Martina, Muhammad Shafique

Recently, Deep Neural Networks (DNNs) have achieved remarkable performances in many applications, while several studies have enhanced their vulnerabilities to malicious attacks. In this paper, we emulate the effects of n…

Targeted Attacks on Timeseries Forecasting

2023-01-27 · Yuvaraj Govindarajulu, Avinash Amballa, Pavan Kulkarni, Manojkumar Parmar

Real-world deep learning models developed for Time Series Forecasting are used in several critical applications ranging from medical devices to the security domain. Many previous works have shown how deep learning models…

Adversarial AttackTime SeriesTime Series AnalysisTime Series Forecasting