学习受控随机微分方程
Learning Controlled Stochastic Differential Equations
AI总结:
针对多维非线性受控随机微分方程,提出核方法拟合漂移与扩散项,证明有限样本误差界并通过数值实验验证,提供开源Python实现。
摘要:
我们研究学习受控随机微分方程(SDEs)的问题,其形式为$dX_t = b(t,X_t,u_t)\\,dt + \sigma(t,X_t,u_t)\\,dW_t$,其中漂移项和扩散项非线性依赖于时间、状态和控制值。从轨迹数据中,我们旨在估计系数,使得该系数诱导的密度流能够复现观测动力学的密度流。数据包含从有限维族中采样得到的若干控制量,且每个控制量对应多条在固定时间范围内离散时刻观测到的独立轨迹;这些控制量是观测输入,而非学习者选择的决策。针对多维非线性受控SDEs,我们提出一种核方法:该方法先估计每个采样控制量对应的密度流,再通过Fokker-Planck方程匹配估计的密度流,拟合漂移项$b$和扩散矩阵$a=\sigma\sigma^\top$。在Sobolev正则性假设下,我们证明了学习得到的SDE与真实SDE的密度流之间误差的有限样本界,该误差在控制量、时间和状态上以$L^2$范数度量。这些界量化了误差如何随采样控制量数量$K$的增加而减小,其收敛速率由状态维度、控制参数化维度以及Sobolev正则性决定。我们还推导了控制量一致的保证条件和针对尾部敏感量的CVaR型界。数值实验验证了该方法,且提供了开源Python实现。
英文摘要:
We study the problem of learning controlled stochastic differential equations (SDEs) \[ dX_t = b(t,X_t,u_t)\,dt + σ(t,X_t,u_t)\,dW_t, \] whose drift and diffusion depend nonlinearly on time, state, and control values. From trajectory data, we aim to estimate coefficients whose induced density flows reproduce those of the observed dynamics. The data consist of several controls sampled from a finite-dimensional family and, for each control, multiple independent trajectories observed at discrete times over a fixed horizon. The controls are observed inputs, not learner-selected decisions. We propose a kernel method for multidimensional nonlinear controlled SDEs. The method estimates the density flow for each sampled control, then fits the drift $b$ and diffusion matrix $a=σσ^\top$ by matching the estimated flows through the Fokker-Planck equation. Under Sobolev regularity assumptions, we prove finite-sample bounds on the error between the density flows of the learned and true SDEs, measured in $L^2$ over controls, time, and state. The bounds quantify how the error decreases with the number $K$ of sampled controls, with rates determined by the state and control-parametrization dimensions and Sobolev regularity. We further derive uniform-in-control guarantees and CVaR-type bounds for tail-sensitive quantities. Numerical experiments illustrate the method, and an open-source Python implementation is provided.
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。