Simulated annealing for fitting linear combinations of Gaussians to data

DOI10.1007/S006070070001MaRDI QIDQ5928221zbMATH OpenOpenAlexFDO

Authors J. R. Parker

Publication date 5 March 2002

Published in Computing (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/s006070070001

zbMATH Keywords

convergence accelerationdata fittingdata modellingGaussianoptimizationsimulated annealingtest samples

Mathematics Subject Classification ID

Algorithms for approximation of functions (65D15) Stochastic programming (90C15)

Review text

In this interesting paper the author presents a number of techniques for accelerating convergence when using simulated annealing for fitting linear combinations of Gaussians to data. The author's suggestions are tested on a sample of known Gaussian combinations and are compared for accuracy and resource consumption. A single `best set' of techniques is found which gives good results on the test samples and on empirical data. A well known fact is demonstrated again, i.e. that providing a good initial parameter estimate is essential to finding a good fit in a reasonable amount of time. The presented approach seems to be promissing.

Recommendations

• Very fast simulated re-annealing
• scientific article; zbMATH DE number 906531
• scientific article; zbMATH DE number 906529
• Simulated annealing: Practice versus theory
• A Bayesian Approach to Simulated Annealing

Cited in

(3)

• Comparison of algorithmic and machine learning approaches for the automatic fitting of Gaussian peaks
• Simulated annealing in compound Gaussian random fields (image processing)
• scientific article; zbMATH DE number 2116107 (Why is no real title available?)

This page was built for publication: Simulated annealing for fitting linear combinations of Gaussians to data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5928221)