A Kernel-Incremental Out-of-Limit Learning MachineAND Differential Multi-Species Grey Wolf Hybrid Optimization Method

Abstract

The invention belongs to the technical field of data analysis and discloses a kernel incremental transfinite learning machine and a differential multi-population gray wolf mixed optimization method. Aiming at the problem that the kernel incremental transfinite learning machine (KI-ELM) has the redundant nodes with low learning efficiency and poor accuracy; At first, that invention utilize the differential evolution algorithm and the multi-population grey wolf optimization algorithm to propose a hybrid intelligent optimization algorithm--the differential multi-population grey wolf optimizationalgorithm, optimizes the node parameter of the hidden layer, and determine the effective node quantity, so as to reduce the network complexity and improve the learning efficiency of the network; Secondly, the depth structure is introduced into the kernel incremental transfinite learning machine, and the input data is extracted layer by layer to realize the high-dimensional mapping classification of the data and improve the classification accuracy and generalization performance of the algorithm. The simulation experiment results show that the hybrid intelligent depth kernel incremental transfinite learning machine provided by the invention has good prediction accuracy and generalization ability, and the network structure is more compact.

Description

technical field [0001] The invention belongs to the technical field of data analysis, and in particular relates to a kernel-incremental extreme learning machine and a differential multi-population gray wolf mixed optimization method. Background technique [0002] At present, the existing technology commonly used in the industry is as follows: the artificial neural network analyzes the data through the abstract simulation of the biological neuron network, so as to realize the functions of data classification, system identification, function approximation and numerical estimation. However, the traditional Single Hidden Layer Feedforward Neural Networks (SLFNS) has low training efficiency and weak learning ability, and all parameters of the network need to be updated in the learning algorithm. An extreme learning machine (Extreme Learning Machine, ELM) algorithm, compared with the traditional neural network, the hidden layer node parameters in ELM are randomly generated, avoidi...

AI Technical Summary

Problems solved by technology

[0003] In summary, the existing problems in the prior art are: the existing Kernel Incremental Extreme Learning Machine (KI-ELM) has redundant nodes with low learning efficiency and poor accuracy.

Images