Riemannian Continuous Normalizing Flows

Riemannian Continuous Normalizing Flows
复制标题

黎曼连续归一化流

DOI:
--
复制
发表时间:
2020
期刊:
Neural Information Processing Systems
影响因子:
--
通讯作者:
Maximilian Nickel
Maximilian Nickel
中科院分区:
--
文献类型:
--
作者:
Emile Mathieu;Maximilian Nickel

文献摘要

参考文献

被引文献

相似文献

归一化流已经显示出以计算上易于处理的方式对灵活概率分布进行建模的巨大前景。然而,虽然数据通常在黎曼流形(例如球体、鸟居和双曲空间)上自然地描述,但大多数归一化流隐含地假设平坦的几何形状,这使得它们在这些情况下要么被错误指定,要么不适合。为了克服这个问题,我们引入了黎曼连续归一化流,该模型通过将流定义为常微分方程的解,允许平滑流形上灵活概率测度的参数化。我们表明,与标准流程或之前引入的预测流程相比,这种方法可以对合成数据和真实世界数据产生重大改进。
Normalizing flows have shown great promise for modelling flexible probability distributions in a computationally tractable way. However, whilst data is often naturally described on Riemannian manifolds such as spheres, torii, and hyperbolic spaces, most normalizing flows implicitly assume a flat geometry, making them either misspecified or ill-suited in these situations. To overcome this problem, we introduce Riemannian continuous normalizing flows, a model which admits the parametrization of flexible probability measures on smooth manifolds by defining flows as the solution to ordinary differential equations. We show that this approach can lead to substantial improvements on both synthetic and real-world data when compared to standard flows or previously introduced projected flows.
DOI: --
发表时间: 2021
期刊: 2020 Advances in Neural Information Processing Systems (NeurIPS 2020
影响因子: --
作者:
Lou, Aaron;Lim, Derek;Katsman, Isay;Huang, Leo;Jiang, Qingxuan;Lim, Ser Nam
通讯作者: Lim, Ser Nam
DOI: --
发表时间: 2020-02
期刊: --
影响因子: --
作者:
Danilo Jimenez Rezende;G. Papamakarios;S. Racanière;M. S. Albergo;G. Kanwar;P. Shanahan;Kyle Cranmer
通讯作者: Danilo Jimenez Rezende;G. Papamakarios;S. Racanière;M. S. Albergo;G. Kanwar;P. Shanahan;Kyle Cranmer
可解释的多项式神经常微分方程。
DOI: 10.1063/5.0130803
发表时间: 2023
期刊: Chaos (Woodbury, N.Y.)
影响因子: --
作者:
Fronk,Colby;Petzold,Linda
通讯作者: Petzold,Linda
DOI: 10.1137/09074721x
发表时间: 2009-01-01
影响因子: 1.5
作者:
Al-Mohy, Awad H.;Higham, Nicholas J.
通讯作者: Higham, Nicholas J.