arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.08561cs.LG
收藏

神经元常微分方程元求解器

Meta-Solver for Neural Ordinary Differential Equations

Skolkovo Institute of Science and Technology(斯科尔科沃科学技术研究院)

Julia Gusak, Alexandr Katrutsa, Talgat Daulbaev, Andrzej Cichocki, Ivan Oseledets

摘要:

训练神经常微分方程(ODE)的传统方法是固定ODE求解器,然后学习神经网络的权重以优化目标损失函数。然而,这种方法是针对特定的离散化方法及其属性定制的,这对于所选择的应用可能不是最佳的,并且产生对给定求解器的过拟合。在我们的论文中,我们研究了求解器空间的可变性如何提高神经常微分方程的性能。我们考虑一个家庭的龙格-库塔方法参数不超过两个标量变量。基于求解器的属性,我们提出了一种方法来减少神经ODE过拟合到预定义的求解器,以及一个标准来评估这种行为。此外,我们表明,正确选择求解器参数化可以显着影响神经ODE模型的鲁棒性对抗攻击。最近的研究表明,神经ODE在鲁棒性方面优于传统的CNN。我们的工作表明,模型的鲁棒性可以进一步提高优化求解器的选择,为给定的任务。复制我们实验的源代码可在https://github.com/juliagusak/neural-ode-metasolver上获得。

英文摘要:

A conventional approach to train neural ordinary differential equations (ODEs) is to fix an ODE solver and then learn the neural network's weights to optimize a target loss function. However, such an approach is tailored for a specific discretization method and its properties, which may not be optimal for the selected application and yield the overfitting to the given solver. In our paper, we investigate how the variability in solvers' space can improve neural ODEs performance. We consider a family of Runge-Kutta methods that are parameterized by no more than two scalar variables. Based on the solvers' properties, we propose an approach to decrease neural ODEs overfitting to the pre-defined solver, along with a criterion to evaluate such behaviour. Moreover, we show that the right choice of solver parameterization can significantly affect neural ODEs models in terms of robustness to adversarial attacks. Recently it was shown that neural ODEs demonstrate superiority over conventional CNNs in terms of robustness. Our work demonstrates that the model robustness can be further improved by optimizing solver choice for a given task. The source code to reproduce our experiments is available at https://github.com/juliagusak/neural-ode-metasolver.

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。

↑