200多支团队竞逐太空航天器数据中的隐藏后门，揭示深度预测模型安全漏洞。
Trojan horse hunt in deep forecasting models: Insights from the European Space Agency competition
- 通过竞赛形式设计新型时序预测模型后门检测任务
- 在航天器遥测数据中成功识别出隐蔽触发模式
- 为关键基础设施的模型安全提供可复现评估方案
预测在现代安全关键应用（如空间操作）中至关重要。然而，深度预测模型的广泛应用引入了后门攻击的新风险：通过在训练数据或模型权重中隐藏后门，在测试时以特定触发模式激活，导致模型输出被操纵。本文介绍的「后门猎手」数据科学竞赛中，200余支队伍需从深时序预测模型中识别航天器遥测数据里的隐藏触发信号。我们提出新颖的任务设定、基准数据集、评估协议，并总结最优解法。进一步归纳了时序预测模型中后门触发识别的关键洞察与研究方向。所有材料均公开于官方竞赛页面 https://www.kaggle.com/competitions/trojan-horse-hunt-in-space。
原文摘要 · Abstract (English)
Forecasting plays a crucial role in modern safety-critical applications, such as space operations. However, the increasing use of deep forecasting models introduces a new security risk of trojan horse attacks, carried out by hiding a backdoor in the training data or directly in the model weights. Once implanted, the backdoor is activated by a specific trigger pattern at test time, causing the model to produce manipulated predictions. We focus on this issue in our \textit{Trojan Horse Hunt} data science competition, where more than 200 teams faced the task of identifying triggers hidden in deep forecasting models for spacecraft telemetry. We describe the novel task formulation, benchmark set, evaluation protocol, and best solutions from the competition. We further summarize key insights and research directions for effective identification of triggers in time series forecasting models. All materials are publicly available on the official competition webpage https://www.kaggle.com/competitions/trojan-horse-hunt-in-space.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。