Optimization method and device

Abstract

The invention provides an optimization method and device, and the method comprises the steps: S1, obtaining a loss function and a termination condition of iteration; S2, determining an observation function of the loss function according to the loss function; S3, determining an iteration function of to-be-optimized parameters according to the observation function, wherein iteration function comprises a random disturbance synchronization vector; S4, carrying out the initialization of the iteration function; S5, determining a random disturbance synchronization vector of the k-th iteration; S6, determining an iteration function of the k-th iteration according to the random disturbance synchronization vector of the k-th iteration and the k-th to-be-optimized parameter; S7, determining the (k+1)-th to-be-optimized parameter according to the iteration function of the k-th iteration; S8, judging whether the termination condition of iteration is met or not: outputting the (k+1)-th to-be-optimized parameter if the termination condition of iteration is met, or else assigning k+1 to k, and returning to step S5, k is an integer which is greater than or equal to zero. The method can reach a better optimization effect in the limited operation cost.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to an optimization method and device. Background technique [0002] In various production processes, many complex optimization problems are often encountered, and often have a certain degree of randomness. How to solve these optimization problems has been paid more and more attention. [0003] In the prior art, there are many schemes for realizing optimization, but these schemes generally require relatively high operating costs to achieve good results, and the optimization effects within limited operating costs are relatively poor. Contents of the invention [0004] The embodiment of the present invention provides an optimization method and device, which can achieve a better optimization effect within a limited operating cost. [0005] On the one hand, an embodiment of the present invention provides an optimization method, including: [0006] S1: Obtain the loss function and t...

AI Technical Summary

Problems solved by technology

[0003] In the prior art, there are many optimization schemes, but these schemes generally require high operating costs to achieve good results, and the optimization effect within the limited operating costs is poor

Images