Simulated Annealing: Rigorous Finite-Time Guarantees for Optimization on Continuous Domains
Abstract
Simulated annealing is a popular method for approaching the solution of a global optimization problem. Existing results on its performance apply to discrete com- binatorial optimization where the optimization variables can assume only a finite set of possible values. We introduce a new general formulation of simulated an- nealing which allows one to guarantee finite-time performance in the optimiza- tion of functions of continuous variables. The results hold universally for any optimization problem on a bounded domain and establish a connection between simulated annealing and up-to-date theory of convergence of Markov chain Monte Carlo methods on continuous domains. This work is inspired by the concept of finite-time learning with known accuracy and confidence developed in statistical learning theory.
Cite
Text
Lecchini-visintini et al. "Simulated Annealing: Rigorous Finite-Time Guarantees for Optimization on Continuous Domains." Neural Information Processing Systems, 2007.Markdown
[Lecchini-visintini et al. "Simulated Annealing: Rigorous Finite-Time Guarantees for Optimization on Continuous Domains." Neural Information Processing Systems, 2007.](https://mlanthology.org/neurips/2007/lecchinivisintini2007neurips-simulated/)BibTeX
@inproceedings{lecchinivisintini2007neurips-simulated,
title = {{Simulated Annealing: Rigorous Finite-Time Guarantees for Optimization on Continuous Domains}},
author = {Lecchini-visintini, Andrea and Lygeros, John and Maciejowski, Jan},
booktitle = {Neural Information Processing Systems},
year = {2007},
pages = {865-872},
url = {https://mlanthology.org/neurips/2007/lecchinivisintini2007neurips-simulated/}
}