UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Analytical Study of Momentum-Based Acceleration Methods in Paradigmatic High-Dimensional Non-Convex Problems

Sarao Mannelli, Stefano; Urbani, Pierfrancesco; (2021) Analytical Study of Momentum-Based Acceleration Methods in Paradigmatic High-Dimensional Non-Convex Problems. In: NeurIPS Proceedings. NeurIPS Green open access

[thumbnail of NeurIPS-2021-analytical-study-of-momentum-based-acceleration-methods-in-paradigmatic-high-dimensional-non-convex-problems-Paper.pdf]
Preview
Text
NeurIPS-2021-analytical-study-of-momentum-based-acceleration-methods-in-paradigmatic-high-dimensional-non-convex-problems-Paper.pdf - Published Version

Download (843kB) | Preview

Abstract

The optimization step in many machine learning problems rarely relies on vanilla gradient descent but it is common practice to use momentum-based accelerated methods. Despite these algorithms being widely applied to arbitrary loss functions, their behaviour in generically non-convex, high dimensional landscapes is poorly understood.In this work, we use dynamical mean field theory techniques to describe analytically the average dynamics of these methods in a prototypical non-convex model: the (spiked) matrix-tensor model. We derive a closed set of equations that describe the behaviour of heavy-ball momentum and Nesterov acceleration in the infinite dimensional limit. By numerical integration of these equations, we observe that these methods speed up the dynamics but do not improve the algorithmic threshold with respect to gradient descent in the spiked model.

Type: Proceedings paper
Title: Analytical Study of Momentum-Based Acceleration Methods in Paradigmatic High-Dimensional Non-Convex Problems
Event: Annual Conference on Neural Information Processing Systems 2021, NeurIPS 2021
Open access status: An open access version is available from UCL Discovery
Publisher version: https://proceedings.neurips.cc/paper/2021/hash/019...
Language: English
Additional information: This version is the version of record. For information on re-use, please refer to the publisher’s terms and conditions.
UCL classification: UCL
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Life Sciences
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Life Sciences > Gatsby Computational Neurosci Unit
URI: https://discovery.ucl.ac.uk/id/eprint/10166774
Downloads since deposit
0Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item