对抗动力系统刻画数据驱动学习何时成功或失败
Adversarial dynamical systems characterize when data-driven learning succeeds or fails
AI总结:
通过对抗动力系统界定数据驱动学习的可行性边界,在Koopman算子学习中给出收敛性与不可能性结果,并在北极海冰预测中超越现有模型。
摘要:
许多系统抗拒解析建模,使得数据驱动的动力学推断变得重要。然而,数据驱动方法可能无法收敛或泛化,留下一个核心问题:何时能从数据中可靠地学习系统行为,何时这种学习不可能?我们通过对抗动力系统来回答这个问题,以识别可访问与不可访问区域之间的边界。在Koopman算子学习中,这是一个通过线性谱对象表示非线性动力学的前沿框架,我们设计了最优的数据驱动谱算法,在物理系统中广泛出现的条件下具有收敛性和认证保证。这为Koopman算子近似提供了收敛理论,并解决了Koopman谱分析中一个长期存在的开放问题。相反,通过构造对抗系统,我们证明了匹配的不可能性结果:没有这些条件,无论数据质量如何,任何单序列极限过程都无法保证学习。这些结果清晰地刻画了数据驱动谱学习何时能够成功以及何时必然失败。我们在振荡器、混沌流体流动和北极海冰浓度预测上验证了该框架。在后者中,我们揭示了北极海冰消退的隐藏模式,提供了具有地理误差界限的长期预测,并以显著更低的计算成本超越了最先进的动力学和深度学习模型,实现了在标准CPU上的实时部署。
英文摘要:
Many systems resist analytical modeling, making data-driven inference of dynamics important. Yet data-driven methods can fail to converge or generalize, leaving open a central question: When can system behavior be learned reliably from data, and when is such learning impossible? We answer this question using adversarial dynamical systems to identify the boundary between accessible and inaccessible regimes. In Koopman operator learning, a leading framework for representing nonlinear dynamics through linear spectral objects, we design optimal data-driven spectral algorithms with convergence and certification guarantees under conditions arising broadly in physical systems. This yields a convergence theory for Koopman-operator approximations and resolves a longstanding open problem in Koopman spectral analysis. Conversely, by constructing adversarial systems, we prove matching impossibility results: without these conditions, no single-sequence limiting procedure can guarantee learning, regardless of data quality. These results sharply characterize when data-driven spectral learning can succeed and when it must fail. We validate the framework on oscillators, chaotic fluid flows and Arctic sea ice concentration forecasting. In the latter, we uncover hidden modes of Arctic sea ice decline, deliver long-range forecasts with geographic error bounds, and outperform state-of-the-art dynamical and deep learning models at substantially lower computational cost, enabling real-time deployment on standard CPUs.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。