Accelerated gradient methods for geodesically convex optimization: Tractable algorithms and convergence analysis

J Kim, I Yang - International Conference on Machine …, 2022 - proceedings.mlr.press
We propose computationally tractable accelerated first-order methods for Riemannian
optimization, extending the Nesterov accelerated gradient (NAG) method. For both …

Negative curvature obstructs acceleration for strongly geodesically convex optimization, even with exact first-order oracles

C Criscitiello, N Boumal - Conference on Learning Theory, 2022 - proceedings.mlr.press
Hamilton and Moitra (2021) showed that, in certain regimes, it is not possible to accelerate
Riemannian gradient descent in the hyperbolic plane if we restrict ourselves to algorithms …

Momentum stiefel optimizer, with applications to suitably-orthogonal attention, and optimal transport

L Kong, Y Wang, M Tao - arXiv preprint arXiv:2205.14173, 2022 - arxiv.org
The problem of optimization on Stiefel manifold, ie, minimizing functions of (not necessarily
square) matrices that satisfy orthogonality constraints, has been extensively studied. Yet, a …

Negative curvature obstructs acceleration for strongly geodesically convex optimization, even with exact first-order oracles

C Criscitiello, N Boumal - arXiv preprint arXiv:2111.13263, 2021 - arxiv.org
Hamilton and Moitra (2021) showed that, in certain regimes, it is not possible to accelerate
Riemannian gradient descent in the hyperbolic plane if we restrict ourselves to algorithms …

Riemannian accelerated gradient methods via extrapolation

A Han, B Mishra, P Jawanpuria… - … Conference on Artificial …, 2023 - proceedings.mlr.press
In this paper, we propose a convergence acceleration scheme for general Riemannian
optimization problems by extrapolating iterates on manifolds. We show that when the …

Accelerated optimization on Riemannian manifolds via discrete constrained variational integrators

V Duruisseaux, M Leok - Journal of Nonlinear Science, 2022 - Springer
A variational formulation for accelerated optimization on normed vector spaces was recently
introduced in Wibisono et al.(PNAS 113: E7351–E7358, 2016), and later generalized to the …

Adaptive Hamiltonian variational integrators and applications to symplectic accelerated optimization

V Duruisseaux, J Schmitt, M Leok - SIAM Journal on Scientific Computing, 2021 - SIAM
It is well known that symplectic integrators lose their near energy preservation properties
when variable time-steps are used. The most common approach to combining adaptive time …

Practical perspectives on symplectic accelerated optimization

V Duruisseaux, M Leok - Optimization Methods and Software, 2023 - Taylor & Francis
Geometric numerical integration has recently been exploited to design symplectic
accelerated optimization algorithms by simulating the Bregman Lagrangian and Hamiltonian …

Time-adaptive Lagrangian variational integrators for accelerated optimization on manifolds

V Duruisseaux, M Leok - arXiv preprint arXiv:2201.03774, 2022 - arxiv.org
A variational framework for accelerated optimization was recently introduced on normed
vector spaces and Riemannian manifolds in Wibisono et al.(2016) and Duruisseaux and …

Variational symplectic accelerated optimization on Lie groups

T Lee, M Tao, M Leok - … 60th IEEE Conference on Decision and …, 2021 - ieeexplore.ieee.org
There has been significant interest in generalizations of the Nesterov accelerated gradient
descent algorithm due to its improved performance guarantee compared to the standard …