Gibbs sampling of complex-valued distributions

作     者：L. L. Salcedo 

作者机构：Departamento de Física Atómica Molecular y Nuclear and Instituto Carlos I de Física Teórica y Computacional Universidad de Granada E-18071 Granada Spain 

出 版 物：《Physical Review D》 (物理学评论D辑：粒子、场、重力与宇宙学)

年 卷 期：2016年第94卷第7期

页      面：074503-074503页

核心收录：

学科分类：07[理学] 070201[理学-理论物理] 0702[理学-物理学] 

基　　金：Centro de Servicios de Informtica y Redes de Comunicaciones Ministerio de Economía y Competitividad, MINECO Universidad de Granada, UGR Agencia de Innovación y Desarrollo de Andalucía, IDEA European Regional Development Fund, FEDER 

主　　题：Lattice field theory Heatbath algorithm Monte Carlo methods 

摘      要：A new technique is explored for the Monte Carlo sampling of complex-valued distributions. The method is based on a heat bath approach where the conditional probability is replaced by a positive representation of it on the complex plane. Efficient ways to construct such representations are also introduced. The performance of the algorithm is tested on small and large lattices with a λϕ4 theory with quadratic nearest-neighbor complex coupling. The method works for moderate complex couplings, reproducing reweighting and complex Langevin results and fulfilling various Schwinger-Dyson relations.