基于隐式模板 ODE-nets 的实现与(逆修正)误差分析
Implementation and (Inverse Modified) Error Analysis for implicitly-templated ODE-nets
AI总结:
研究基于隐式求解器模板的 ODE-nets 学习未知动力学,通过逆修正误差分析建立超参数选择理论基础,提出自适应调整隐式解迭代次数的算法以加速训练并保持精度。
摘要:
我们专注于使用基于隐式数值初值问题求解器模板的 ODE-nets 从数据中学习未知动力学。首先,为了便于解释,我们使用展开的隐式格式对 ODE-nets 进行逆修正误差分析。结果表明,使用展开的隐式格式训练 ODE-net 会返回逆修正微分方程(IMDE)的近似解。此外,我们为训练此类 ODE-nets 时的超参数选择建立了理论基础,而当前策略通常将 ODE-nets 的数值积分视为黑盒。因此,我们制定了一种自适应算法,该算法在训练过程中监控误差水平并调整(展开的)隐式解迭代的次数,从而使展开近似的误差小于当前的学习损失。这有助于在保持准确性的同时加速训练。进行了几项数值实验,以证明所提出算法相较于非自适应展开的优势,并验证了理论分析。我们还注意到,这种方法自然地允许在方程中并入部分已知的物理项,从而产生了所谓的“灰盒”识别。
英文摘要:
We focus on learning unknown dynamics from data using ODE-nets templated on implicit numerical initial value problem solvers. First, we perform Inverse Modified error analysis of the ODE-nets using unrolled implicit schemes for ease of interpretation. It is shown that training an ODE-net using an unrolled implicit scheme returns a close approximation of an Inverse Modified Differential Equation (IMDE). In addition, we establish a theoretical basis for hyper-parameter selection when training such ODE-nets, whereas current strategies usually treat numerical integration of ODE-nets as a black box. We thus formulate an adaptive algorithm which monitors the level of error and adapts the number of (unrolled) implicit solution iterations during the training process, so that the error of the unrolled approximation is less than the current learning loss. This helps accelerate training, while maintaining accuracy. Several numerical experiments are performed to demonstrate the advantages of the proposed algorithm compared to nonadaptive unrollings, and validate the theoretical analysis. We also note that this approach naturally allows for incorporating partially known physical terms in the equations, giving rise to what is termed ``gray box" identification.
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。