Riemannian Deep Learning: Modules, Networks, and Geometries
Abstract
Deep neural networks on manifold-valued representations have attracted growing interest, but many basic components remain tied to specific manifolds, rely on Euclidean approximations, or require costly and numerically fragile geometric operations. This thesis develops a unified framework for Riemannian deep learning from three complementary perspectives: reusable neural modules, manifold-specific network architectures, and the design of underlying geometries. It generalizes batch normalization from Euclidean spaces and individual manifolds to broad classes of Lie groups and gyrogroups, and extends multinomial logistic regression from Euclidean space to SPD manifolds and then to general Riemannian manifolds. It further develops neural networks for several important geometric representations, including an unconstrained model of hyperbolic space, Busemann-based hyperbolic learning, and full-rank correlation matrices. Finally, it introduces adaptive and computationally efficient Riemannian metrics on SPD manifolds, including learnable Log-Euclidean geometries and fast, stable Cholesky-based geometries. The proposed methods are supported by theoretical analysis and validated through numerical experiments and applications in vision, signal processing, graph learning, and genomics.
Related
- Minimax rates for learning spectral Barron functions by deep ReLU neural networks
- Bayesian Neural Networks via Dirichlet Process Priors on the Data-Generating Distribution for Randomized Value Functions in Deep Reinforcement Learning
- Efficient Joint Trajectory Optimization and Resource Allocation for Secure UAV Networks: A Deep Reinforcement Learning Approach
- Deep Learning-Based Multi-User Communication Design for Dense IoT Networks: Interference-Aware Finite-Blocklength Communication and Preliminary MIMO Extensions
- Investigating the Early Signatures of Feature Learning in Deep Linear Networks
BibTeX
@misc{chen2026riemannian,
title = {{Riemannian Deep Learning: Modules, Networks, and Geometries}},
author = {Ziheng Chen},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2607.19305},
eprint = {2607.19305},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2607.19305},
note = {CC BY-NC-SA 4.0},
}
RIS
TY - UNPB
TI - Riemannian Deep Learning: Modules, Networks, and Geometries
AU - Ziheng Chen
PY - 2026
DA - 2026/10/29
PB - arXiv
DO - 10.48550/arxiv.2607.19305
UR - https://arxiv.org/abs/2607.19305
AB - Deep neural networks on manifold-valued representations have attracted growing interest, but many basic components remain tied to specific manifolds, rely on Euclidean approximations, or require costly and numerically fragile geometric operations. This thesis develops a unified framework for Riemannian deep learning from three complementary perspectives: reusable neural modules, manifold-specific network architectures, and the design of underlying geometries. It generalizes batch normalization from Euclidean spaces and individual manifolds to broad classes of Lie groups and gyrogroups, and extends multinomial logistic regression from Euclidean space to SPD manifolds and then to general Riemannian manifolds. It further develops neural networks for several important geometric representations, including an unconstrained model of hyperbolic space, Busemann-based hyperbolic learning, and full-rank correlation matrices. Finally, it introduces adaptive and computationally efficient Riemannian metrics on SPD manifolds, including learnable Log-Euclidean geometries and fast, stable Cholesky-based geometries. The proposed methods are supported by theoretical analysis and validated through numerical experiments and applications in vision, signal processing, graph learning, and genomics.
ER -