Riemannian Adaptive Optimization Methods


METADATA ONLY
Loading...

Date

2023-05

Publication Type

Conference Paper

ETH Bibliography

yes

Citations

Altmetric
METADATA ONLY

Data

Rights / License

Abstract

Several first order stochastic optimization methods commonly used in the Euclidean domain such as stochastic gradient descent (SGD), accelerated gradient descent or variance reduced methods have already been adapted to certain Riemannian settings. However, some of the most popular of these optimization tools - namely Adam , Adagrad and the more recent Amsgrad - remain to be generalized to Riemannian manifolds. We discuss the difficulty of generalizing such adaptive schemes to the most agnostic Riemannian setting, and then provide algorithms and convergence proofs for geodesically convex objectives in the particular case of a product of Riemannian manifolds, in which adaptivity is implemented across manifolds in the cartesian product. Our generalization is tight in the sense that choosing the Euclidean space as Riemannian manifold yields the same algorithms and regret bounds as those that were already known for the standard algorithms. Experimentally, we show faster convergence and to a lower train loss value for Riemannian adaptive methods over their corresponding baselines on the realistic task of embedding the WordNet taxonomy in the Poincare ball.

Permanent link

Publication status

published

External links

Editor

Book title

International Conference on Learning Representations (ICLR 2019)

Journal / series

Volume

9

Pages / Article No.

6384 - 6399

Publisher

Curran

Event

7th International Conference on Learning Representations (ICLR 2019)

Edition / version

Methods

Software

Geographic location

Date collected

Date created

Subject

Organisational unit

09462 - Hofmann, Thomas / Hofmann, Thomas

Notes

Conference lecture held on May 8, 2019.

Funding

Related publications and datasets