Extreme learning machine method for improving artificial bee colony optimization

Abstract

The invention discloses an extreme learning machine method for improving artificial bee colony optimization. The method is characterized by comprising the following steps: 1, generating initial solutions for SN individuals; 2, carrying out global optimization on a connection weight Omega and a threshold b of an extreme learning machine; 3, carrying out local optimization on the connection weight Omega and the threshold b of the extreme learning machine; 4, if food source information is not updated in a certain time, converting employee bees into investigation bees and returning to the step 1 to re-initiate the individuals; and 5, extracting the connection weight Omega and the threshold b of the extreme learning machine from the optimum individual, and carrying out verification by using a test set. The method disclosed in the invention can be used for better overcoming the defect that the traditional extreme learning machines are relatively bad in results when being applied to classification and regression; and compared with the traditional extreme learning machines and SaE-ELM algorithms, the method has relatively strong robustness and can be used for effectively improving the results of classification and regression.

Description

technical field [0001] The invention belongs to the technical field of artificial intelligence and relates to an improved extreme learning machine method, in particular to an improved extreme learning machine method for artificial bee colony optimization. Background technique [0002] Artificial Neural Networks (ANN) is an algorithmic mathematical model that imitates the behavioral characteristics of biological neural networks and performs distributed parallel computing. Among them, Single-hidden Layer Feedforward Neural Networks (SLFNs) have been widely used in many fields because of their good learning ability. However, since most of the traditional feedforward neural networks use the gradient descent method to correct the value of hidden layer nodes, it is prone to disadvantages such as slow training speed, easy to fall into local minimum points, and more parameters need to be set. In recent years, a new type of feedforward neural network - Extreme Learning Machine (Extr...

AI Technical Summary

Problems solved by technology

"Zhu Q Y, Qin A K, Suganthan P N, et al.Evolutionary ex-treme learningmachine[J].Pattern recognition,2005,38(10):1759-1763." proposed an evolutionary extreme learning machine (E-ELM) , the algorithm uses the differential evolution algorithm to optimize the hidden layer node parameters of the ELM, thereby improving the performance of the ELM, but more parameters need to be set, and the experimental process is complicated; "Cao J, Lin Z, Huang G B.Self-adaptiveevolutionary ex-treme learning machine[J].Neural processing letters,2012,36(3):285-305."A self-adaptive evolutionary extreme learning machine (SaE-ELM) is proposed, and the algorithm combines an adaptive evolutionary The algorithm is combined with the extreme learning machine, and the hidden layer nodes are optimized on the basis of setting fewer parameters, which improves the accuracy and stability of the extreme learning machine in regression and classification problems, but the algorithm uses The time is too long and the practicability is poor; "Wang Jie, Bi Haoyang. An extreme learning machine based on particle swarm optimization [J]. Journal of Zhengzhou University: Science Edition, 2013,45(1):100-104." proposed An extreme learning machine based on particle swarm optimization (PSO-ELM), using the particle swarm optimization algorithm to optimize and select the input layer weight and hidden layer deviation of the extreme learning machine, so as to obtain an optimal network, but the algorithm is only in The function fitting has achieved good results, but the effect is not good in practical applications; "Lin Meijin, Luo Fei, Su Caihong, et al. A new hybrid intelligent extreme learning machine [J]. Control and Decision, 2015, 30(06):1078-1084." Combining the differential evolution algorithm and the particle swarm optimization algorithm, referring to the memetic evolution mechanism of the leapfrog algorithm, a hybrid intelligent optimization algorithm (DEPSO-ELM) is proposed for parameter optimization. The extreme learning machine algorithm obtains the output weights of SLFNs, but the algorithm relies too much on experimental data and has poor robustness

Images