Riemannian Deep Learning: Modules, Networks, and Geometries

[Submitted on 21 Jul 2026 (v1), last revised 9 Aug 2026 (this version, v3)]

View PDF

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.

Submission history

From: Ziheng Chen [view email]
[v1] Tue, 21 Jul 2026 17:17:35 UTC (23,055 KB)
[v2] Wed, 22 Jul 2026 21:13:10 UTC (23,055 KB)
[v3] Sun, 9 Aug 2026 18:07:22 UTC (23,055 KB)