Accurate Sampling with Noisy Forces from Approximate Computing
Computational Physics
2020-04-29 v3 Statistical Mechanics
Numerical Analysis
Numerical Analysis
Chemical Physics
Abstract
In scientific computing, the acceleration of atomistic computer simulations by means of custom hardware is finding ever growing application. A major limitation, however, is that the high efficiency in terms of performance and low power consumption entails the massive usage of low-precision computing units. Here, based on the approximate computing paradigm, we present an algorithmic method to rigorously compensate for numerical inaccuracies due to low-accuracy arithmetic operations, yet still obtaining exact expectation values using a properly modified Langevin-type equation.
Cite
@article{arxiv.1907.08497,
title = {Accurate Sampling with Noisy Forces from Approximate Computing},
author = {Varadarajan Rengaraj and Michael Lass and Christian Plessl and Thomas D. Kühne},
journal= {arXiv preprint arXiv:1907.08497},
year = {2020}
}