Communication: optimal parameters for basin-hopping global optimization based on Tsallis statistics.
Change log
Authors
Shang, C
Wales, DJ
Abstract
A fundamental problem associated with global optimization is the large free energy barrier for the corresponding solid-solid phase transitions for systems with multi-funnel energy landscapes. To address this issue we consider the Tsallis weight instead of the Boltzmann weight to define the acceptance ratio for basin-hopping global optimization. Benchmarks for atomic clusters show that using the optimal Tsallis weight can improve the efficiency by roughly a factor of two. We present a theory that connects the optimal parameters for the Tsallis weighting, and demonstrate that the predictions are verified for each of the test cases.
Description
Keywords
0103 Numerical and Computational Mathematics
Journal Title
J Chem Phys
Conference Name
Journal ISSN
0021-9606
1089-7690
1089-7690
Volume Title
141
Publisher
AIP Publishing
Publisher DOI
Sponsorship
European Research Council (267369)
This work was supported by the ERC and the EPSRC.