Parameter optimization and feature tuning method and system for machine learning

Abstract

The invention provides a parameter optimization and feature tuning method for machine learning. The method includes the following steps that a plurality of parameter sets are generated randomly; iterative optimization based on EnKF is conducted on the parameter sets; performance evaluation is conducted on the optimized parameter sets, a set pool and a supplement parameter set are obtained according to evaluation results, and the performance of the parameter sets in the set pool is higher than the performance of the parameter sets in the supplement parameter set; iterative optimization based on the EnKF and performance evaluation are conducted on the parameter sets in the set pool and the parameter sets in the supplement parameter set again, and accordingly an optimal parameter set is obtained. By the adoption of the method, the computational efficiency for processing the computed results of parameter optimization can be improved, and universality is high. The invention further provides a parameter optimization and feature tuning system for machine learning.

Description

Technical field [0001] The present invention relates to the technical field of parameter optimization of machine learning, in particular to a method and system for parameter optimization and feature optimization of machine learning. Background technique [0002] For general machine learning algorithms, the performance of the model mainly depends on its parameter configuration. Models generated with different parameter combinations often have great performance differences. Parameter optimization is a random optimization problem, and its randomness is mainly reflected in the fact that the training data and test data used to generate the model contain limited samples and cannot reflect the whole, and the parameter space is based on an unknown joint distribution function. The basic definition of the problem is as follows: Given a training data set X T , Where X T Based on the unknown data distribution G, the goal of parameter optimization is to find a parameter combination θ of mach...

AI Technical Summary

Problems solved by technology

[0008] The scaling factor of the feature is used as a class of parameters to complete the feature tuning, and the optimization together with the parameters of the machine learning algorithm will lead to a large number of parameters

However, there is still a lack of fast, accurate, general-purpose, and effective techniques for machine learning parameter optimization, especially for high-dimensional continuous parameter space optimization.

Images