Simulated Annealing using Hybrid Monte Carlo

TL;DR

This paper introduces a hybrid Monte Carlo-based variant of simulated annealing that enhances optimization efficiency for high-dimensional differentiable functions by enabling more effective search strategies and faster annealing schedules.

Contribution

It presents a novel hybrid Monte Carlo approach integrated into simulated annealing, improving performance in multivariate optimization tasks.

Findings

Enhanced search efficiency in high-dimensional spaces

Faster convergence compared to traditional simulated annealing

Better performance in multivariate differentiable functions

Abstract

We propose a variant of the Simulated Annealing method for optimization in the multivariate analysis of differentiable functions. The method uses global actualizations via the Hybrid Monte Carlo algorithm in their generalized version for the proposal of new configurations. We show how this choice can improve upon the performance of simulated annealing methods (mainly when the number of variables is large) by allowing a more effective searching scheme and a faster annealing schedule.