Warped geometric information on the optimisation of Euclidean functions

TL;DR

This paper introduces a Riemannian geometric approach with warped metrics to optimize Euclidean functions more efficiently, leveraging analytical geodesic approximations and retraction maps to improve convergence in high-dimensional spaces.

Contribution

It proposes a novel optimization method using warped Riemannian metrics and third-order geodesic approximations, enhancing convergence over traditional Euclidean methods.

Findings

Outperforms standard Euclidean gradient methods in convergence speed.

Uses third-order Taylor approximations for efficient geodesic computation.

Empirically validated on challenging optimization benchmarks.

Abstract

We consider the fundamental task of optimising a real-valued function defined in a potentially high-dimensional Euclidean space, such as the loss function in many machine-learning tasks or the logarithm of the probability distribution in statistical inference. We use Riemannian geometry notions to redefine the optimisation problem of a function on the Euclidean space to a Riemannian manifold with a warped metric, and then find the function's optimum along this manifold. The warped metric chosen for the search domain induces a computational friendly metric-tensor for which optimal search directions associated with geodesic curves on the manifold becomes easier to compute. Performing optimization along geodesics is known to be generally infeasible, yet we show that in this specific manifold we can analytically derive Taylor approximations up to third-order. In general these approximations to the geodesic curve will not lie on the manifold, however we construct suitable retraction maps to pull them back onto the manifold. Therefore, we can efficiently optimize along the approximate geodesic curves. We cover the related theory, describe a practical optimization algorithm and empirically evaluate it on a collection of challenging optimisation benchmarks. Our proposed algorithm, using 3rd-order approximation of geodesics, tends to outperform standard Euclidean gradient-based counterparts in term of number of iterations until convergence.