【预测模型-DELM分类】基于哈里斯鹰算法改进深度学习极限学习机实现数据分类附matlab代码

✅作者简介：热爱科研的Matlab仿真开发者，修心和技术同步精进，matlab项目合作可私信。

🍎个人主页：Matlab科研工作室

🍊个人信条：格物致知。

更多Matlab仿真内容点击👇

智能优化算法  神经网络预测 雷达通信  无线传感器

信号处理 图像处理 路径规划 元胞自动机 无人机

⛄ 内容介绍

​人工时间的最大缺点是训练太长，因为它在应用神经网络的时间范围内，持续不断地限制神经网络，最大限度地限制学习机（Extreme Learning Machine）大量的噪声噪声，或者当输入数据时的维度算法非常高时，极限学习时的综合性能会受到极大的影响。进行空间映射时的有效对数据维的维度的预测，因此我们认为利用深度学习的预测精度来最大学习机的特性，可以很好地改善极限学习机的特性。 本文采用哈里斯鹰算法的进一步优化DELM超参数，仿真结果改进，预测精度更高。

⛄ 部分代码

% Developed in MATLAB R2013b

% Source codes demo version 1.0

% _____________________________________________________

% Main paper:

% Harris hawks optimization: Algorithm and applications

% Ali Asghar Heidari, Seyedali Mirjalili, Hossam Faris, Ibrahim Aljarah, Majdi Mafarja, Huiling Chen

% Future Generation Computer Systems, 

% DOI: https://doi.org/10.1016/j.future.2019.02.028

% https://www.sciencedirect.com/science/article/pii/S0167739X18313530

% _____________________________________________________

% You can run the HHO code online at codeocean.com  https://doi.org/10.24433/CO.1455672.v1

% You can find the HHO code at https://github.com/aliasghar68/Harris-hawks-optimization-Algorithm-and-applications-.git

% _____________________________________________________

%  Author, inventor and programmer: Ali Asghar Heidari,

%  PhD research intern, Department of Computer Science, School of Computing, National University of Singapore, Singapore

%  Exceptionally Talented Ph. DC funded by Iran's National Elites Foundation (INEF), University of Tehran

%  03-03-2019

%  Researchgate: https://www.researchgate.net/profile/Ali_Asghar_Heidari

%  e-Mail: as_heidari@ut.ac.ir, aliasghar68@gmail.com,

%  e-Mail (Singapore): aliasgha@comp.nus.edu.sg, t0917038@u.nus.edu

% _____________________________________________________

%  Co-author and Advisor: Seyedali Mirjalili

%

%         e-Mail: ali.mirjalili@gmail.com

%                 seyedali.mirjalili@griffithuni.edu.au

%

%       Homepage: http://www.alimirjalili.com

% _____________________________________________________

%  Co-authors: Hossam Faris, Ibrahim Aljarah, Majdi Mafarja, and Hui-Ling Chen

%       Homepage: http://www.evo-ml.com/2019/03/02/hho/

% _____________________________________________________

%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

% Harris's hawk optimizer: In this algorithm, Harris' hawks try to catch the rabbit.

% T: maximum iterations, N: populatoin size, CNVG: Convergence curve

% To run HHO: [Rabbit_Energy,Rabbit_Location,CNVG]=HHO(N,T,lb,ub,dim,fobj)

function [Rabbit_Energy,Rabbit_Location,CNVG]=HHO(N,T,lb,ub,dim,fobj)

disp('HHO is now tackling your problem')

tic

% initialize the location and Energy of the rabbit

Rabbit_Location=zeros(1,dim);

Rabbit_Energy=inf;

%Initialize the locations of Harris' hawks

X=initialization(N,dim,ub,lb);

CNVG=zeros(1,T);

t=0; % Loop counter

while t

    for i=1:size(X,1)

        % Check boundries

        FU=X(i,:)>ub;FL=X(i,:)

        % fitness of locations

        fitness=fobj(X(i,:));

        % Update the location of Rabbit

        if fitness

            Rabbit_Energy=fitness;

            Rabbit_Location=X(i,:);

        end

    end

    

    E1=2*(1-(t/T)); % factor to show the decreaing energy of rabbit

    % Update the location of Harris' hawks

    for i=1:size(X,1)

        E0=2*rand()-1; %-1

        Escaping_Energy=E1*(E0);  % escaping energy of rabbit

        

        if abs(Escaping_Energy)>=1

            %% Exploration:

            % Harris' hawks perch randomly based on 2 strategy:

            

            q=rand();

            rand_Hawk_index = floor(N*rand()+1);

            X_rand = X(rand_Hawk_index, :);

            if q<0.5

                % perch based on other family members

                X(i,:)=X_rand-rand()*abs(X_rand-2*rand()*X(i,:));

            elseif q>=0.5

                % perch on a random tall tree (random site inside group's home range)

                X(i,:)=(Rabbit_Location(1,:)-mean(X))-rand()*((ub-lb)*rand+lb);

            end

            

        elseif abs(Escaping_Energy)<1

            %% Exploitation:

            % Attacking the rabbit using 4 strategies regarding the behavior of the rabbit

            

            %% phase 1: surprise pounce (seven kills)

            % surprise pounce (seven kills): multiple, short rapid dives by different hawks

            

            r=rand(); % probablity of each event

            

            if r>=0.5 && abs(Escaping_Energy)<0.5 % Hard besiege

                X(i,:)=(Rabbit_Location)-Escaping_Energy*abs(Rabbit_Location-X(i,:));

            end

            

            if r>=0.5 && abs(Escaping_Energy)>=0.5  % Soft besiege

                Jump_strength=2*(1-rand()); % random jump strength of the rabbit

                X(i,:)=(Rabbit_Location-X(i,:))-Escaping_Energy*abs(Jump_strength*Rabbit_Location-X(i,:));

            end

            

            %% phase 2: performing team rapid dives (leapfrog movements)

            if r<0.5 && abs(Escaping_Energy)>=0.5, % Soft besiege % rabbit try to escape by many zigzag deceptive motions

                

                Jump_strength=2*(1-rand());

                X1=Rabbit_Location-Escaping_Energy*abs(Jump_strength*Rabbit_Location-X(i,:));

                

                if fobj(X1)

                    X(i,:)=X1;

                else % hawks perform levy-based short rapid dives around the rabbit

                    X2=Rabbit_Location-Escaping_Energy*abs(Jump_strength*Rabbit_Location-X(i,:))+rand(1,dim).*Levy(dim);

                    if (fobj(X2)

                        X(i,:)=X2;

                    end

                end

            end

            

            if r<0.5 && abs(Escaping_Energy)<0.5, % Hard besiege % rabbit try to escape by many zigzag deceptive motions

                % hawks try to decrease their average location with the rabbit

                Jump_strength=2*(1-rand());

                X1=Rabbit_Location-Escaping_Energy*abs(Jump_strength*Rabbit_Location-mean(X));

                

                if fobj(X1)

                    X(i,:)=X1;

                else % Perform levy-based short rapid dives around the rabbit

                    X2=Rabbit_Location-Escaping_Energy*abs(Jump_strength*Rabbit_Location-mean(X))+rand(1,dim).*Levy(dim);

                    if (fobj(X2)

                        X(i,:)=X2;

                    end

                end

            end

            %%

        end

    end

    t=t+1;

    CNVG(t)=Rabbit_Energy;

%    Print the progress every 100 iterations

%    if mod(t,100)==0

%        display(['At iteration ', num2str(t), ' the best fitness is ', num2str(Rabbit_Energy)]);

%    end

end

toc

end

% ___________________________________

function o=Levy(d)

beta=1.5;

sigma=(gamma(1+beta)*sin(pi*beta/2)/(gamma((1+beta)/2)*beta*2^((beta-1)/2)))^(1/beta);

u=randn(1,d)*sigma;v=randn(1,d);step=u./abs(v).^(1/beta);

o=step;

end

⛄ 运行结果

⛄ 参考文献

[1]吴丁杰, 温立书. 一种基于哈里斯鹰算法优化的核极限学习机[J]. 信息通信, 2021(034-011).

❤️ 关注我领取海量matlab电子书和数学建模资料

❤️部分理论引用网络文献，若有侵权联系博主删除