Multi-scale quantum harmonic oscillator optimization method with inherent parallel ability

Abstract

The invention belongs to the technical field of function optimization. In order to solve the technical problems that an existing function optimization algorithm is low in parallel efficiency and the optimal solution position is prone to loss, a multi-scale quantum harmonic oscillator optimization method with the inherent parallel ability comprises the steps that firstly, the value of sigmamin is set, the value of sigmas is set, and K first random numbers are decimated randomly in the domain of definition of f(x); secondly, K second random numbers are generated through sigmas and the first random numbers according to normal distribution; thirdly, the K first random numbers are substituted into f(x) to obtain K first function values, and the K second random numbers are substituted into f(x) to obtain K second function values; fourthly, for any xi and xi', if f(xi') is smaller than f(xi), xi is substituted with xi', if substitution exists, the second step is executed, and if no substitution exists, the fifth step is executed; fifthly, Xworst is substituted with Xmean, and the sixth step is executed; sixthly, if sigmak is larger than sigmas, the second step is executed, otherwise, the value of sigmas is reduced, and then the seventh step is executed; seventhly, if sigmas is larger than sigmamin, the second step is executed, and otherwise, f(xbest) and xbest serve as results to be output.

Description

technical field [0001] The invention belongs to the technical field of function optimization artificial intelligence, and specifically relates to a multi-scale quantum harmonic oscillator optimization method with inherent parallel capability. Background technique [0002] With the rapid development of computer artificial intelligence, more and more problems are solved through cloud computing platform and big data analysis. Among the many problems that need to be solved by the cloud computing platform, it is often necessary to find the optimal solution through convergence, such as the shortest time to go from one place to another, the lowest energy consumption, the best solution, etc. Generally speaking, any All problems can be converted into mathematical objective functions, and the problems can be solved by solving the optimal solution of the objective function. However, since the process of finding the optimal solution requires a large amount of calculation, it usually ne...

AI Technical Summary

Problems solved by technology

[0004] (1) The energy level stabilization process is the core iterative process of the MQHOA algorithm, but since each iteration needs to calculate the standard deviation σ of k Gaussian sampling center positions k , so that the sampling operation of each Gaussian sampling area cannot be performed independently. This problem makes the algorithm need frequent message passing to calculate the standard deviation when running on a parallel cluster, and the parallel efficiency is very low.

This defect prevents the algorithm from effectively using high-performance computers based on large-scale cluster technology to optimize ultra-high-dimensional complex functions.

[0005] (2) From the perspective of the physical model, the energy level drop criterion of the current algorithm is the standard deviation σ of the k Gaussian sampling center positions between two energy level stabilization operations k difference (Δσ k =|σ k -σ k ′|) is smaller than the current scale, this criterion cannot guarantee the complete stability of the algorithm in the metastable state, making the algorithm insufficient iterations in each local optimal region, which may cause the loss of the optimal solution position

Images