Open Access

Table C.1

Overview and comparison of different algorithms

Algorithm Nmodels Advantage Disadvantage
Grid scan with local minimization 107 Considers more than only one solution, can be reused for other sources, few controllable parameters Computationally expensive, drastic increase in computing cost with the increase of model parameters
Genetic algorithm 105−106 Able to cover large part of parameter space, overcomes local minima Only one solution, sensitive to setting of algorithm parameters
CMA-ES 105 Able to cover large part of parameter space, overcomes local minima, self-adaption of algorithm parameters Only one solution
Minuit (simplex + migrad) 103 Computationally inexpensive, few controllable parameters May be sensitive to local minima, only one solution, sensitive to the choice of initial point
MCMC (ensemble sampler) 105 Provides parameter distributions Sensitive to the choice of initial point and algorithm parameters

Notes. Nmodels is the total number of generated models in the selected algorithm.

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.