Convergence acceleration in machine learning potentials for atomistic simulations
نویسندگان
چکیده
Machine learning potentials (MLPs) for atomistic simulations have an enormous prospective impact on materials modeling, offering orders of magnitude speedup over density functional theory without appreciably sacrificing accuracy material property prediction.
منابع مشابه
Deriving effective mesoscale potentials from atomistic simulations
We demonstrate how an iterative method for potential inversion from distribution functions developed for simple liquid systems can be generalized to polymer systems. It uses the differences in the potentials of mean force between the distribution functions generated from a guessed potential and the true distribution functions to improve the effective potential successively. The optimization alg...
متن کاملInteratomic potentials for atomistic simulations of the Ti-Al system
Semiempirical interatomic potentials have been developed for Al, a-Ti, and g-TiAl within the embedded atom method ~EAM! formalism by fitting to a large database of experimental as well as ab initio data. The ab initio calculations were performed by the linearized augmented plane wave ~LAPW! method within the density functional theory to obtain the equations of state for a number of crystal stru...
متن کاملRedox reactions with empirical potentials: atomistic battery discharge simulations.
Batteries are pivotal components in overcoming some of today's greatest technological challenges. Yet to date there is no self-consistent atomistic description of a complete battery. We take first steps toward modeling of a battery as a whole microscopically. Our focus lies on phenomena occurring at the electrode-electrolyte interface which are not easily studied with other methods. We use the ...
متن کاملConvergence of reinforcement learning algorithms and acceleration of learning.
The techniques of reinforcement learning have been gaining increasing popularity recently. However, the question of their convergence rate is still open. We consider the problem of choosing the learning steps alpha(n), and their relation with discount gamma and exploration degree epsilon. Appropriate choices of these parameters may drastically influence the convergence rate of the techniques. F...
متن کاملConditions for Convergence in Regularized Machine Learning Objectives
Analysis of the convergence rates of modern convex optimization algorithms can be achived through binary means: analysis of emperical convergence, or analysis of theoretical convergence. These two pathways of capturing information diverge in efficacy when moving to the world of distributed computing, due to the introduction of non-intuitive, non-linear slowdowns associated with broadcasting, an...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Digital discovery
سال: 2022
ISSN: ['2635-098X']
DOI: https://doi.org/10.1039/d1dd00005e