Riemannian Neural SDE: Learning Stochastic Representations on Manifolds
Abstract
In recent years, the neural stochastic differential equation (NSDE) has gained attention in modeling stochastic representations, while NSDE brings a great success in various types of applications. However, it typically loses the expressivity when the data representation is manifold-valued. To overcome such an issue, we suggest a principled way to express the stochastic representation with the Riemannian neural SDE (RNSDE), which extends the conventional Euclidean NSDE. Empirical results on the density estimation on manifolds show that the proposed method significantly outperforms baseline methods.
Cite
Text
Park et al. "Riemannian Neural SDE: Learning Stochastic Representations on Manifolds." ICLR 2022 Workshops: GTRL, 2022.Markdown
[Park et al. "Riemannian Neural SDE: Learning Stochastic Representations on Manifolds." ICLR 2022 Workshops: GTRL, 2022.](https://mlanthology.org/iclrw/2022/park2022iclrw-riemannian/)BibTeX
@inproceedings{park2022iclrw-riemannian,
title = {{Riemannian Neural SDE: Learning Stochastic Representations on Manifolds}},
author = {Park, Sung Woo and Kim, Hyomin and Kim, Hyeseong and Kwon, Junseok},
booktitle = {ICLR 2022 Workshops: GTRL},
year = {2022},
url = {https://mlanthology.org/iclrw/2022/park2022iclrw-riemannian/}
}