arXivDaily arXiv每日学术速递 周一至周五更新
arXiv quant-ph/0205020quant-phcond-mat.dis-nnphysics.chem-ph
收藏

用量子退火方法研究优化问题

Study of Optimization Problems by Quantum Annealing

Tadashi Kadowaki

摘要:

我们将量子涨落引入到优化问题的模拟退火过程中,旨在更快地收敛到最优状态。用横向伊辛模型和旅行商问题(TSP)两个模型对这一思想进行了验证。在伊辛模型中加入横场是引入量子涨落的一种简单方法。与传统方法中的温度类似,横向场的强度被控制为时间的函数。TSP可以用伊辛自旋来描述,因此我们也将同样的技术应用于TSP。我们直接求解了小尺寸系统的含时薛定谔方程,并对大尺寸系统进行了量子蒙特卡罗模拟。与相应的经典(热)方法的结果比较表明,量子方法导致基态的概率大得多,在几乎所有的情况下,如果我们使用相同的退火时间表的控制参数。通过数值模拟,我们还发现量子系统的弛豫时间很短。我们认为这是量子系统中的退火比经典系统具有更好的寻优性能的原因之一。

英文摘要:

We introduce quantum fluctuations into the simulated annealing process of optimization problems, aiming at faster convergence to the optimal state. The idea is tested by the two models, the transverse Ising model and the traveling salesman problem (TSP). Adding the transverse field to the Ising model is a simple way to introduce quantum fluctuations. The strength of the transverse field is controlled as a function of time similarly to the temperature in the conventional method. TSP can be described by the Ising spin, so that we also apply the same technique to TSP. We directory solve the time-dependent Schrödinger equation for small-size systems and perform the quantum Monte Carlo simulation for large-size systems. Comparison with the results of the corresponding classical (thermal) method reveals that the quantum method leads to the ground state with much larger probability in almost all cases if we use the same annealing schedule of the control parameters. We also find that the relaxation time is quite short for quantum systems by numerical simulations. We consider this is one of the reasons why the annealing in quantum systems have a better performance of finding the optimal state in comparison with classical systems.

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。

↑