过于罕见而无法学习：规定的气旋轨迹会降低孟加拉湾海洋模拟器的性能
Too Rare to Learn: Prescribed Cyclone Tracks Degrade a Bay of Bengal Ocean Emulator
- University of Dhaka（达卡大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
该研究发现，在孟加拉湾的神经海洋模拟器中，将气旋轨迹作为规定输入的U-Net模型性能劣于仅海洋模型，因气旋轨迹信号罕见致网络响应错误，替换为无风暴图可提升预报性能。
AI中文摘要：
目前正提出神经海洋模拟器用于受气旋影响的沿海海域的区域预报，一种自然的设计选择是将气旋作为规定输入提供给网络。我们在孟加拉湾对这一选择进行了测试，发现它是有害的。我们从GLORYS12再分析数据中预留了15个完整的气旋，强度范围为65至150节，并比较了两个U-Net模型，除了四个规定的气旋轨迹通道外，这两个模型完全相同。在三个随机种子下，仅海洋模型在每次运行中都优于持续性预报，而受风暴条件约束的模型在每次运行中都劣于持续性预报，两者的技能范围完全不重叠（p=3.1e-5，按风暴配对）。原因是暴露频率而非信号内容：这些通道仅在7.9%的训练日非零，因此它们激活的那一刻就处于分布外。额外的误差出现在规定的风暴足迹内，在推理时将真实气旋图替换为无风暴图，可使每个种子的预留风暴预报提高7.5%至16.4%。该受约束网络已学会对罕见信号做出自信但错误的响应。
英文摘要：
Neural ocean emulators are being proposed for regional forecasting in cyclone-exposed coastal seas, and a natural design choice is to hand the network the cyclone as a prescribed input. We test that choice in the Bay of Bengal and find it harmful. We withhold 15 whole cyclones spanning 65 to 150 kt from GLORYS12 reanalysis and compare two U-Nets that are identical except for four prescribed cyclone-track channels. Across three seeds the ocean-only model beats persistence in every run and the storm-conditioned model loses to it in every run, with the two skill ranges disjoint (p = 3.1e-5, paired across storms). The cause is exposure frequency rather than signal content: the channels are non-zero on only 7.9% of training days, so they are out of distribution the moment they activate. The extra error falls inside the prescribed storm footprint, and replacing the real cyclone map with a no-storm map at inference improves held-out storm forecasts by 7.5 to 16.4% in every seed. The conditioned network has learned a response to a rare signal that is confidently wrong.
