arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 0706.4408cond-mat.dis-nncond-mat.stat-mech
收藏

连续自旋系统基态的遗传嵌入匹配方法

Genetic embedded matching approach to ground states in continuous-spin systems

Martin Weigel

摘要:

由于自由能景观的极其崎岖的结构,自旋玻璃基态的确定是已知的最困难的优化问题之一,发现在最一般的情况下是NP-难的。由于局部(自由)能最小值的特定结构,通用优化策略在这些问题上的表现相对较差,并且已经开发了一些特别定制的优化技术,特别是针对伊辛自旋玻璃和类似的离散系统。在这里,一个有效的优化启发式少得多讨论的情况下,连续自旋的介绍,基于组合的伊辛自旋到连续的旋转器和适当的变种的遗传算法。统计技术,以确保高可靠性,在寻找(数值)准确的基态进行了讨论,该方法是对模拟退火方法的基准。

英文摘要:

Due to an extremely rugged structure of the free energy landscape, the determination of spin-glass ground states is among the hardest known optimization problems, found to be NP-hard in the most general case. Owing to the specific structure of local (free) energy minima, general-purpose optimization strategies perform relatively poorly on these problems, and a number of specially tailored optimization techniques have been developed in particular for the Ising spin glass and similar discrete systems. Here, an efficient optimization heuristic for the much less discussed case of continuous spins is introduced, based on the combination of an embedding of Ising spins into the continuous rotators and an appropriate variant of a genetic algorithm. Statistical techniques for insuring high reliability in finding (numerically) exact ground states are discussed, and the method is benchmarked against the simulated annealing approach.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑