CoolMomentum: a method for stochastic optimization by Langevin dynamics with simulated annealing
Abstract Deep learning applications require global optimization of non-convex objective functions, which have multiple local minima. The same problem is often found in physical simulations and may be resolved by the methods of Langevin dynamics with Simulated Annealing, which is a well-established a...
Main Authors: | Oleksandr Borysenko, Maksym Byshkin |
---|---|
Format: | Article |
Language: | English |
Published: |
Nature Portfolio
2021-05-01
|
Series: | Scientific Reports |
Online Access: | https://doi.org/10.1038/s41598-021-90144-3 |
Similar Items
-
CoolMomentum-SPGD Algorithm for Wavefront Sensor-Less Adaptive Optics Systems
by: Zhiguang Zhang, et al.
Published: (2023-01-01) -
Consistency and fluctuations for stochastic gradient Langevin dynamics
by: Teh, YW, et al.
Published: (2016) -
Benchmarking quantum annealing dynamics: The spin-vector Langevin model
by: David Subires, et al.
Published: (2022-05-01) -
Bayesian learning via stochastic gradient langevin dynamics
by: Welling, M, et al.
Published: (2011) -
Conformational Transitions of Amyloid-β: A Langevin and Generalized Langevin Dynamics Simulation Study
by: Vishal Singh, et al.
Published: (2021-05-01)