发表机构
University of Chicago; University of California, Santa Cruz（芝加哥大学; 加州大学圣克鲁兹分校）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究开发概率深度学习模拟器模拟随机平流层突然增温，通过条件变分自编码器准确再现罕见转变动力学，并在潜在空间中无监督分离出物理可解释的簇，揭示极端事件流形，助力预警系统。
AI 中文摘要
罕见天气状态转变因类别不平衡对数据驱动建模构成挑战。在本研究中，我们开发了一个用于具有状态转变的原型系统——平流层变率的随机Holton-Mass模型的概率深度学习模拟器，并分析了其学习到的潜在空间结构。Holton-Mass模型表现出两个亚稳态状态，即强极涡和弱极涡，由非线性波-平均流相互作用维持，弱随机强迫间歇性地触发这些状态之间的罕见转变，这些转变定性地代表了平流层突然增温事件。我们采用受ResNet启发的条件变分自编码器，具有六层编码器和解码器层以及显式的当前状态条件化，来模拟系统在下一步时间（一天）状态分布的模型。该模拟器准确再现了物理模型的短期动力学、稳态概率分布、状态持续性统计、罕见转变率、转变通勤函数以及转变预期提前时间。除了模拟保真度之外，我们探究了学习到的潜在表示，以理解模型如何内化动力学中潜在的亚稳态结构。对32维潜在空间的主成分分析揭示了清晰且无监督的分离，形成四个物理上可解释的簇，对应于强涡与弱涡状态以及稳定与易转变配置。这种潜在空间中涌现的状态分离对于应用于高维随机系统的深度生成模型来说难以识别。我们的结果表明，精心设计的概率模拟器能够揭示控制极端事件动力学的物理上有意义的流形，可能有助于开发改进的操作性先进预警系统。
英文摘要
Rare weather regime transitions pose a challenge for data-driven modeling due to class imbalance. In this study, we develop a probabilistic deep learning emulator for a prototypical system with regime transitions, the stochastic Holton--Mass model of stratospheric variability, and analyze the structure of its learned latent space. The Holton--Mass model exhibits two metastable regimes, a strong and a weak polar vortex, maintained by nonlinear wave--mean flow interactions, with weak stochastic forcing intermittently triggering rare transitions between these regimes that qualitatively represent SSW events. We employ a ResNet-inspired Conditional Variational Autoencoder with six-layer encoder and decoder layers and explicit current-state conditioning to model the distribution of the system's state at the next time step (one day). The emulator accurately reproduces short-term dynamics, steady-state probability distributions, regime persistence statistics, rare transition rates, the transition committor function, and the transition expected lead time of the physical model. Beyond emulation fidelity, we interrogate the learned latent representation to understand how the model internalizes the underlying metastable structure of the dynamics. Principal Component Analysis of the 32-dimensional latent space reveals a clear and unsupervised separation into four physically interpretable clusters corresponding to strong versus weak vortex regimes and stable versus transition-prone configurations. Such emergent regime separation in latent space is hard to identify for deep generative models applied to high-dimensional stochastic systems. Our results show that carefully designed probabilistic emulators can uncover physically meaningful manifolds governing extreme-event dynamics, potentially aiding the development of improved operational advanced warning systems.
