Swarm-Based Gradient Descent Method for Non-Convex Optimization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lu, Jingcheng, Tadmor, Eitan, Zenginoglu, Anil
Format: Preprint
Published: 2022
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913334612197376
author Lu, Jingcheng
Tadmor, Eitan
Zenginoglu, Anil
author_facet Lu, Jingcheng
Tadmor, Eitan
Zenginoglu, Anil
contents We introduce a new Swarm-Based Gradient Descent (SBGD) method for non-convex optimization. The swarm consists of agents, each is identified with a position, ${\mathbf x}$, and mass, $m$. The key to their dynamics is communication: masses are being transferred from agents at high ground to low(-est) ground. At the same time, agents change positions with step size, $h=h({\mathbf x},m)$, adjusted to their relative mass: heavier agents proceed with small time-steps in the direction of local gradient, while lighter agents take larger time-steps based on a backtracking protocol. Accordingly, the crowd of agents is dynamically divided between `heavier' leaders, expected to approach local minima, and `lighter' explorers. With their large-step protocol, explorers are expected to encounter improved position for the swarm; if they do, then they assume the role of `heavy' swarm leaders and so on. Convergence analysis and numerical simulations in one-, two-, and 20-dimensional benchmarks demonstrate the effectiveness of SBGD as a global optimizer.
format Preprint
id arxiv_https___arxiv_org_abs_2211_17157
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Swarm-Based Gradient Descent Method for Non-Convex Optimization
Lu, Jingcheng
Tadmor, Eitan
Zenginoglu, Anil
Numerical Analysis
Optimization and Control
90C26, 65K10, 92D25
We introduce a new Swarm-Based Gradient Descent (SBGD) method for non-convex optimization. The swarm consists of agents, each is identified with a position, ${\mathbf x}$, and mass, $m$. The key to their dynamics is communication: masses are being transferred from agents at high ground to low(-est) ground. At the same time, agents change positions with step size, $h=h({\mathbf x},m)$, adjusted to their relative mass: heavier agents proceed with small time-steps in the direction of local gradient, while lighter agents take larger time-steps based on a backtracking protocol. Accordingly, the crowd of agents is dynamically divided between `heavier' leaders, expected to approach local minima, and `lighter' explorers. With their large-step protocol, explorers are expected to encounter improved position for the swarm; if they do, then they assume the role of `heavy' swarm leaders and so on. Convergence analysis and numerical simulations in one-, two-, and 20-dimensional benchmarks demonstrate the effectiveness of SBGD as a global optimizer.
title Swarm-Based Gradient Descent Method for Non-Convex Optimization
topic Numerical Analysis
Optimization and Control
90C26, 65K10, 92D25
url https://arxiv.org/abs/2211.17157