arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.01650cs.LG
收藏

基于自同态的常微分方程快速鲁棒积分

Learning ODEs via Diffeomorphisms for Fast and Robust Integration

the University of Sydney(悉尼大学);ETH Zurich(苏黎世联邦理工学院);CSIRO’s Data61(澳大利亚联邦科学与工业研究组织数据61研究所);NVIDIA(英伟达)

Weiming Zhi, Tin Lai, Lionel Ott, Edwin V. Bonilla, Fabio Ramos

摘要:

可微数值积分器的进步使得能够使用梯度下降技术来学习常微分方程(ODE)。在机器学习的背景下,可微求解器是神经ODE(NODE)的核心,NODE是一类具有连续深度而不是离散层的深度学习模型。然而,当从长序列学习常微分方程系统时,或者当系统的解在每个维度上以不同的时间尺度变化时,这些积分器可能是令人不满意的缓慢和不准确的。在本文中,我们提出了另一种方法来学习常微分方程的数据:我们表示的基础常微分方程作为一个向量场,是相关的另一个基本向量场的可微双射,可逆神经网络建模。通过限制基础常微分方程服从积分,我们可以大大加快和提高集成的鲁棒性。我们证明了我们的方法在训练和评估连续神经网络模型,以及在学习基准ODE系统的有效性。我们观察到的改进高达两个数量级时,集成学习的ODE与GPU计算。

英文摘要:

Advances in differentiable numerical integrators have enabled the use of gradient descent techniques to learn ordinary differential equations (ODEs). In the context of machine learning, differentiable solvers are central for Neural ODEs (NODEs), a class of deep learning models with continuous depth, rather than discrete layers. However, these integrators can be unsatisfactorily slow and inaccurate when learning systems of ODEs from long sequences, or when solutions of the system vary at widely different timescales in each dimension. In this paper we propose an alternative approach to learning ODEs from data: we represent the underlying ODE as a vector field that is related to another base vector field by a differentiable bijection, modelled by an invertible neural network. By restricting the base ODE to be amenable to integration, we can drastically speed up and improve the robustness of integration. We demonstrate the efficacy of our method in training and evaluating continuous neural networks models, as well as in learning benchmark ODE systems. We observe improvements of up to two orders of magnitude when integrating learned ODEs with GPUs computation.

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。

↑