Convergence of Hyperbolic Neural Networks Under Riemannian Stochastic Gradient Descent

Convergence of Hyperbolic Neural Networks Under Riemannian Stochastic Gradient Descent
复制标题

黎曼随机梯度下降下双曲神经网络的收敛性

DOI:
10.1007/s42967-023-00302-9
复制
发表时间:
2023
影响因子:
1.6
通讯作者:
Xin, Jack
Xin, Jack
中科院分区:
数学4区
文献类型:
--
作者:
Whiting, Wes;Wang, Bao;Xin, Jack

文献摘要

相似文献

在温和的条件下,证明了一类双曲神经网络回归模型的riemanian梯度下降法在批量梯度下降和随机梯度下降下的收敛性。我们还讨论了亚当算法的黎曼版本。我们展示了这些算法在各种基准上的数值模拟。
We prove, under mild conditions, the convergence of a Riemannian gradient descent method for a hyperbolic neural network regression model, both in batch gradient descent and stochastic gradient descent. We also discuss a Riemannian version of the Adam algorithm. We show numerical simulations of these algorithms on various benchmarks.