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

by   Christopher Criscitiello, et al.

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 which make queries in a (large) bounded domain and which receive gradients and function values corrupted by a (small) amount of noise. We show that acceleration remains unachievable for any deterministic algorithm which receives exact gradient and function-value information (unbounded queries, no noise). Our results hold for the classes of strongly and nonstrongly geodesically convex functions, and for a large class of Hadamard manifolds including hyperbolic spaces and the symmetric space SL(n) / SO(n) of positive definite n × n matrices of determinant one. This cements a surprising gap between the complexity of convex optimization and geodesically convex optimization: for hyperbolic spaces, Riemannian gradient descent is optimal on the class of smooth and and strongly geodesically convex functions, in the regime where the condition number scales with the radius of the optimization domain. The key idea for proving the lower bound consists of perturbing the hard functions of Hamilton and Moitra (2021) with sums of bump functions chosen by a resisting oracle.


page 1

page 2

page 3

page 4


Acceleration in Hyperbolic and Spherical Spaces

We further research on the acceleration phenomenon on Riemannian manifol...

No-go Theorem for Acceleration in the Hyperbolic Plane

In recent years there has been significant effort to adapt the key tools...

Complexity of Highly Parallel Non-Smooth Convex Optimization

A landmark result of non-smooth convex optimization is that gradient des...

No quantum speedup over gradient descent for non-smooth convex optimization

We study the first-order convex optimization problem, where we have blac...

Towards Riemannian Accelerated Gradient Methods

We propose a Riemannian version of Nesterov's Accelerated Gradient algor...

A Riemannian Corollary of Helly's Theorem

We introduce a notion of halfspace for Hadamard manifolds that is natura...

Making Progress Based on False Discoveries

We consider the question of adaptive data analysis within the framework ...