Breaking reversibility accelerates Langevin dynamics for global non-convex optimization
Langevin dynamics (LD) has been proven to be a powerful technique for optimizing a non-
convex objective as an efficient algorithm to find local minima while eventually visiting a
global minimum on longer time-scales. LD is based on the first-order Langevin diffusion
which is reversible in time. We study two variants that are based on non-reversible Langevin
diffusions: the underdamped Langevin dynamics (ULD) and the Langevin dynamics with a
non-symmetric drift (NLD). Adopting the techniques of Tzen, Liang and Raginsky (2018) for …
convex objective as an efficient algorithm to find local minima while eventually visiting a
global minimum on longer time-scales. LD is based on the first-order Langevin diffusion
which is reversible in time. We study two variants that are based on non-reversible Langevin
diffusions: the underdamped Langevin dynamics (ULD) and the Langevin dynamics with a
non-symmetric drift (NLD). Adopting the techniques of Tzen, Liang and Raginsky (2018) for …
[CITATION][C] Breaking Reversibility Accelerates Langevin Dynamics for Global Non-Convex Optimization. arXiv e-prints, page
X Gao, M Gurbuzbalaban, L Zhu - arXiv preprint arXiv:1812.07725, 2018
Showing the best results for this search. See all results