2024/12/23 by Yonglong Ding, Ding, Yonglong
Decision Sciences · #Computational Physics (physics.comp-ph) #Data Analysis #FOS: Physical sciences #Simulation Techniques and Applications #Statistics and Probability (physics.data-an)
paper · pdf · doi:10.48550/arxiv.2501.03248
openalex publication_date 2024/12/23 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
As the particle count escalates, the computational demands of diverse simulation algorithms surge, paralleled by a marked enhancement in accuracy. The question arises whether this heightened precision asymptotically dwindles towards zero or plateaus at a finite constant. To address this, this work introduces an approach that translates infinite systems into finite-node network architectures, providing a rigorous framework for assessing this question. Employing the Monte Carlo algorithm's application to the Ising model as a case study, this paper demonstrate that despite the simulation's extension to an infinite lattice size, a fundamental error bound persists. This work explicitly derive this lower bound on the error, offering a quantitative understanding of the algorithm's limitations in the limit of infinite scale. Furthermore, I extend this methodology to Molecular Dynamics simulations, exemplified through its application to battery systems. This conversion strategy not only underscores the generality of this approach but also highlights its practical significance in guiding the optimization of simulation algorithms. Moreover, it offers insights into estimating micro-level information from macro-level data. The crucial information of Molecular Simulation, namely the potential energy, has been quickly estimated.