Stochastic Decomposition: A New Monte Carlo Algorithm for High-Efficiency Sampling and Applications to Quantum and Classical Problems

南加州大学的麦志浩获得了理论和计算化学项目的支持，继续开发和应用一种基于随机分解算法(SDA)的新型非传统蒙特卡罗方法。采样效率的提高使得对各种复杂系统的统计机制的研究具有极高的计算效率。正在进行的研究包括应用路径积分模拟来研究低温系统中的量子效应，应用凝聚相系统中的量子动力学来研究量子团簇和液体，与膜和表面相关的现象，以及Lennard-Jones液体等稠密流体。具体的应用包括RNA的三级结构和折叠、核小体中的DNA包装、囊泡萌发、融合和相关的膜现象，以及超富里氦和氢纳米团簇。这项工作在增加我们对复杂物理、化学和生物系统中合作行为起源的理解方面产生了更广泛的影响。参与这项研究的学生接受现代计算和模拟方法的高级培训。

Chi Ho H. Mak of the University of Southern California is supported by the Theoretical and Computational Chemistry Program to continue the development and application of a novel and unconventional Monte Carlo method based on the stochastic decomposition algorithm (SDA). Improvements in sampling efficiency allow studies of the statistical mechanics of a variety of complex systems with highly accelerated computational efficiency. Research is underway to include application of path integral simulations to study quantum effects in low-temperature systems, quantum dynamics in condensed-phase systems to study quantum clusters and liquids, membrane- and surface-related phenomena, and dense fluids such as the Lennard-Jones liquid.. Specific applications include the tertiary structure and folding of RNA, DNA packaging in the nucleosome, vesicle budding, fusion and related membrane phenomena, and superfulid helium and hydrogen nanoclusters. This work is having a broader impact in increasing our understanding of the origin of cooperative behavior in complex physical, chemical and biological systems. Students participating in this research receive advanced training in modern methods of computation and simulation.