SALTO: Solver-Anchored Lifted Transfer Operators for Nonlinear Event Inference
作者: Qi Wang, Rylie Foster
分类: cs.LG, math.DS
发布日期: 2026-10-15 (更新: 2026-10-16)
备注: 22 pages, 11 figures, 12 tables. Under review. v2: corrected author affiliation
💡 一句话要点
提出SALTO框架以解决非线性事件推断问题
🎯 匹配领域: 支柱六:视频提取与匹配 (Video Extraction)
关键词: 非线性动力学 事件推断 Koopman理论 伴随方法 有限幅度源 科学决策 动态系统建模
📋 核心要点
- 现有的学习动力学模型在求导时灵敏度不可靠,限制了其在逆问题中的应用。
- SALTO框架通过线性传播子近似非线性动力学,并使用提升空间中的转置进行高效推断。
- 在三个非线性PDE系统中,SALTO显著提高了测量预测的准确性,且计算成本大幅降低。
📝 摘要(中文)
学习的动力学模型在预测轨迹时表现良好,但在求导或转置时产生不可靠的灵敏度,限制了其在逆问题和科学决策中的应用。经典伴随方法提供了精确的局部导数,但单一伴随无法表示系统对有限幅度扰动的非线性响应。本文提出SALTO(Solver-Anchored Lifted Transfer Operators),这是一个基于Koopman的框架,用于推断有限幅度源。在提升空间中,非线性动力学通过线性传播子进行近似,其转置在单次求解中将每个传感器向后传递。该模型通过三项结构约束确保与模拟器的精确局部行为一致。实验表明,SALTO在三个非线性PDE系统中预测有限幅度源的测量误差仅为2-8%,而经典伴随方法的误差为41-165%。
🔬 方法详解
问题定义:本文旨在解决学习动力学模型在逆问题中灵敏度不可靠的问题,现有的经典伴随方法无法有效表示非线性系统对有限幅度扰动的响应。
核心思路:SALTO框架通过将非线性动力学近似为线性传播子,并在提升空间中进行推断,从而实现对有限幅度源的有效推断。该设计使得模型能够在单次求解中高效地预测测量结果。
技术框架:SALTO框架包含三个主要模块:首先是将物理状态动力学锚定到模拟器的线性传播子,其次是中心化学习的可观测量以确保在参考状态下二阶消失,最后是使用块三角提升传播子。
关键创新:SALTO的创新在于其可重用的提升伴随,能够在一次编码传递和内积计算中预测任何候选局部事件的测量结果。这一方法确保了模型在幅度消失时能够还原为模拟器的切线动力学和经典伴随。
关键设计:模型设计中采用了三项结构约束,确保了与模拟器的精确一致性,且在参数设置上优化了学习的可观测量,以便在有限幅度下进行有效的修正。具体的损失函数和网络结构细节在论文中有详细描述。
🖼️ 关键图片
📊 实验亮点
在三个非线性PDE系统的实验中,SALTO能够将有限幅度源的测量预测误差控制在2-8%之间,而经典伴随方法的误差范围为41-165%。此外,SALTO在源位置和强度的恢复上几乎与求解器的反演结果相当,同时以更低的计算成本实现了对两个交互源的联合后验采样。
🎯 应用场景
SALTO框架在科学决策、逆问题求解和动态系统建模等领域具有广泛的应用潜力。其高效的推断能力和准确性使其能够在气候建模、流体动力学等复杂系统中发挥重要作用,未来可能推动相关领域的研究进展。
📄 摘要(原文)
Learned dynamical models can accurately predict trajectories while producing unreliable sensitivities when differentiated or transposed, limiting their use in inverse problems and scientific decision-making. Classical adjoints provide exact local derivatives, but a single adjoint cannot represent the nonlinear response of a system to a family of finite-amplitude perturbations. We introduce SALTO (Solver-Anchored Lifted Transfer Operators), a Koopman-based framework for inferring finite-amplitude sources. In the lifted space, the nonlinear dynamics are approximated by a linear propagator, and its transpose carries each sensor backward in a single solve. The result is a reusable lifted adjoint that predicts the measurement produced by any candidate localized event with one encoder pass and an inner product. To give the model the simulator's exact local behavior, SALTO imposes three structural constraints: it anchors the physical-state dynamics to the simulator's linearized propagator, centers the learned observables so that they vanish to second order at the reference state, and uses a block-triangular lifted propagator. These constraints guarantee that the model reduces to the simulator's tangent dynamics and classical adjoint as the amplitude vanishes, while the learned observables carry only the finite-amplitude correction. On three nonlinear PDE systems, SALTO predicts the measurements of finite-amplitude sources to within 2-8% where the classical adjoint is in error by 41-165%, recovers source position and intensity from sparse sensors nearly as accurately as inversion through the solver, and samples the joint posterior of two interacting sources, matching the solver's posterior on reaction-diffusion at a twentieth of its cost. One trained model serves new sensor layouts and measurement times with a single backward solve.