机器学习笔记 十一：最强大的学习算法之支持向量机（一）

1. 高斯核函数（非线性决策边界）

$$\begin{array}{r}{K}_{\text{gaussian }}\left(x^{(i)},x^{(j)}\right)=\exp \left(-\frac{{\Vert x^{(i)}-x^{(j)}\Vert }^2}{2{\sigma }^2}\right)=\exp \left(-\frac{{\sum }_{k=1}^n{\left(x_k^{(i)}-x_k^{(j)}\right)}^2}{2{\sigma }^2}\right)\end{array}$$

1.1 matlab实现

% 高斯核函数，用于计算样本之间的相似度
function sim = gaussianKernel(x1, x2, sigma)

% 使x1和x2都为列向量
x1 = x1(:); x2 = x2(:);

sim = 0;

% 相似程度比较（通过向量实现）
sim = exp(-sum(x1-x2).^2)/(2*sigma^2);

end

1.2 核函数的图像：

 由图可知，$\sigma$影响着图像曲面的陡峭程度，$\sigma$越大图像越平缓，值越小图像越陡峭。

1.3 地标的选取

 若训练集中有n个样本，则会选择n个地标。
 核函数的主要目的就是求得两个样本之前的相似度，在非线性分类问题中（或者复杂的分类问题中），自动创建新的特征。这些新特征是建立在原有特征与训练集的其他特征之间的距离的基础上产生的，即自己利用自己。而这些其他特征就叫做地标（landmarks），类似于反距离加权法（IDW）。

2. 获取最优的参数（C、$\sigma$）

# 选择最优的C、sigma，确保代价函数最小
function [C, sigma] = dataset3Params(X, y, Xval, yval)


% You need to return the following variables correctly.
C = 1;
sigma = 0.3;

% 给出8个数字：0.01, 0.03, 0.1, 0.3, 1, 3, 10, 30
% 分别从中选取两个数字作为C和sigma
% 所以最终会产生8*8=64种模型
% 并从中选择代价最小的一种模型店C和sigma
vec_const = [0.01;0.03;0.1;0.3;1;3;10;30];
C_optimal = Inf; % 无限大
sigma_optimal = Inf;
err_min = Inf;

for i = 1:length(vec_const),
  C_curr = vec_const(i);
  for j = 1:length(vec_const),
    sigma_curr = vec_const(j);
    model = svmTrain(X,y,C_curr,@(x1,x2) gaussianKernel(x1,x2,sigma_curr));
    predictions = svmPredict(model,Xval);
    pre_err = mean(double(predictions ~= yval));
    if(pre_err < err_min)
      C_optimal = C_curr;
      sigma_optimal = sigma_curr;
      err_min = pre_err;
    endif;
  endfor;
end;

C = C_optimal;
sigma = sigma_optimal;

end

结果展示：

2.1 svmTrain.m

模型训练

function [model] = svmTrain(X, Y, C, kernelFunction, ...
                            tol, max_passes)
%SVMTRAIN Trains an SVM classifier using a simplified version of the SMO 
%algorithm. 
%   [model] = SVMTRAIN(X, Y, C, kernelFunction, tol, max_passes) trains an
%   SVM classifier and returns trained model. X is the matrix of training 
%   examples.  Each row is a training example, and the jth column holds the 
%   jth feature.  Y is a column matrix containing 1 for positive examples 
%   and 0 for negative examples.  C is the standard SVM regularization 
%   parameter.  tol is a tolerance value used for determining equality of 
%   floating point numbers. max_passes controls the number of iterations
%   over the dataset (without changes to alpha) before the algorithm quits.
%
% Note: This is a simplified version of the SMO algorithm for training
%       SVMs. In practice, if you want to train an SVM classifier, we
%       recommend using an optimized package such as:  
%
%           LIBSVM   (http://www.csie.ntu.edu.tw/~cjlin/libsvm/)
%           SVMLight (http://svmlight.joachims.org/)
%
%

if ~exist('tol', 'var') || isempty(tol)
    tol = 1e-3;
end

if ~exist('max_passes', 'var') || isempty(max_passes)
    max_passes = 5;
end

% Data parameters
m = size(X, 1);
n = size(X, 2);

% Map 0 to -1
Y(Y==0) = -1;

% Variables
alphas = zeros(m, 1);
b = 0;
E = zeros(m, 1);
passes = 0;
eta = 0;
L = 0;
H = 0;

% Pre-compute the Kernel Matrix since our dataset is small
% (in practice, optimized SVM packages that handle large datasets
%  gracefully will _not_ do this)
% 
% We have implemented optimized vectorized version of the Kernels here so
% that the svm training will run faster.
if strcmp(func2str(kernelFunction), 'linearKernel')
    % Vectorized computation for the Linear Kernel
    % This is equivalent to computing the kernel on every pair of examples
    K = X*X';
elseif strfind(func2str(kernelFunction), 'gaussianKernel')
    % Vectorized RBF Kernel
    % This is equivalent to computing the kernel on every pair of examples
    X2 = sum(X.^2, 2);
    K = bsxfun(@plus, X2, bsxfun(@plus, X2', - 2 * (X * X')));
    K = kernelFunction(1, 0) .^ K;
else
    % Pre-compute the Kernel Matrix
    % The following can be slow due to the lack of vectorization
    K = zeros(m);
    for i = 1:m
        for j = i:m
             K(i,j) = kernelFunction(X(i,:)', X(j,:)');
             K(j,i) = K(i,j); %the matrix is symmetric
        end
    end
end

% Train
fprintf('\nTraining ...');
dots = 12;
while passes < max_passes,
            
    num_changed_alphas = 0;
    for i = 1:m,
        
        % Calculate Ei = f(x(i)) - y(i) using (2). 
        % E(i) = b + sum (X(i, :) * (repmat(alphas.*Y,1,n).*X)') - Y(i);
        E(i) = b + sum (alphas.*Y.*K(:,i)) - Y(i);
        
        if ((Y(i)*E(i) < -tol && alphas(i) < C) || (Y(i)*E(i) > tol && alphas(i) > 0)),
            
            % In practice, there are many heuristics one can use to select
            % the i and j. In this simplified code, we select them randomly.
            j = ceil(m * rand());
            while j == i,  % Make sure i \neq j
                j = ceil(m * rand());
            end

            % Calculate Ej = f(x(j)) - y(j) using (2).
            E(j) = b + sum (alphas.*Y.*K(:,j)) - Y(j);

            % Save old alphas
            alpha_i_old = alphas(i);
            alpha_j_old = alphas(j);
            
            % Compute L and H by (10) or (11). 
            if (Y(i) == Y(j)),
                L = max(0, alphas(j) + alphas(i) - C);
                H = min(C, alphas(j) + alphas(i));
            else
                L = max(0, alphas(j) - alphas(i));
                H = min(C, C + alphas(j) - alphas(i));
            end
           
            if (L == H),
                % continue to next i. 
                continue;
            end

            % Compute eta by (14).
            eta = 2 * K(i,j) - K(i,i) - K(j,j);
            if (eta >= 0),
                % continue to next i. 
                continue;
            end
            
            % Compute and clip new value for alpha j using (12) and (15).
            alphas(j) = alphas(j) - (Y(j) * (E(i) - E(j))) / eta;
            
            % Clip
            alphas(j) = min (H, alphas(j));
            alphas(j) = max (L, alphas(j));
            
            % Check if change in alpha is significant
            if (abs(alphas(j) - alpha_j_old) < tol),
                % continue to next i. 
                % replace anyway
                alphas(j) = alpha_j_old;
                continue;
            end
            
            % Determine value for alpha i using (16). 
            alphas(i) = alphas(i) + Y(i)*Y(j)*(alpha_j_old - alphas(j));
            
            % Compute b1 and b2 using (17) and (18) respectively. 
            b1 = b - E(i) ...
                 - Y(i) * (alphas(i) - alpha_i_old) *  K(i,j)' ...
                 - Y(j) * (alphas(j) - alpha_j_old) *  K(i,j)';
            b2 = b - E(j) ...
                 - Y(i) * (alphas(i) - alpha_i_old) *  K(i,j)' ...
                 - Y(j) * (alphas(j) - alpha_j_old) *  K(j,j)';

            % Compute b by (19). 
            if (0 < alphas(i) && alphas(i) < C),
                b = b1;
            elseif (0 < alphas(j) && alphas(j) < C),
                b = b2;
            else
                b = (b1+b2)/2;
            end

            num_changed_alphas = num_changed_alphas + 1;

        end
        
    end
    
    if (num_changed_alphas == 0),
        passes = passes + 1;
    else
        passes = 0;
    end

    fprintf('.');
    dots = dots + 1;
    if dots > 78
        dots = 0;
        fprintf('\n');
    end
    if exist('OCTAVE_VERSION')
        fflush(stdout);
    end
end
fprintf(' Done! \n\n');

% Save the model
idx = alphas > 0;
model.X= X(idx,:);
model.y= Y(idx);
model.kernelFunction = kernelFunction;
model.b= b;
model.alphas= alphas(idx);
model.w = ((alphas.*Y)'*X)';

end

2.2 svmPrediction.m

代价预测函数

function pred = svmPredict(model, X)
%SVMPREDICT returns a vector of predictions using a trained SVM model
%(svmTrain). 
%   pred = SVMPREDICT(model, X) returns a vector of predictions using a 
%   trained SVM model (svmTrain). X is a mxn matrix where there each 
%   example is a row. model is a svm model returned from svmTrain.
%   predictions pred is a m x 1 column of predictions of {0, 1} values.
%

% Check if we are getting a column vector, if so, then assume that we only
% need to do prediction for a single example
if (size(X, 2) == 1)
    % Examples should be in rows
    X = X';
end

% Dataset 
m = size(X, 1);
p = zeros(m, 1);
pred = zeros(m, 1);

if strcmp(func2str(model.kernelFunction), 'linearKernel')
    % We can use the weights and bias directly if working with the 
    % linear kernel
    p = X * model.w + model.b;
elseif strfind(func2str(model.kernelFunction), 'gaussianKernel')
    % Vectorized RBF Kernel
    % This is equivalent to computing the kernel on every pair of examples
    X1 = sum(X.^2, 2);
    X2 = sum(model.X.^2, 2)';
    K = bsxfun(@plus, X1, bsxfun(@plus, X2, - 2 * X * model.X'));
    K = model.kernelFunction(1, 0) .^ K;
    K = bsxfun(@times, model.y', K);
    K = bsxfun(@times, model.alphas', K);
    p = sum(K, 2);
else
    % Other Non-linear kernel
    for i = 1:m
        prediction = 0;
        for j = 1:size(model.X, 1)
            prediction = prediction + ...
                model.alphas(j) * model.y(j) * ...
                model.kernelFunction(X(i,:)', model.X(j,:)');
        end
        p(i) = prediction + model.b;
    end
end

% Convert predictions into 0 / 1
pred(p >= 0) =  1;
pred(p <  0) =  0;

end

 参数C的作用类似于逻辑回归中的$\lambda$，表示函数的一个权重（惩罚程度），其中，C越大，SVM将对所有的数据点进行尽可能准确的分类（这样就会产生过拟合的情况）：

当C设置的不大时，将会得到图像：

 这才是我们希望得到的图像，所以，当C不是特别大时，可以忽略掉一些异常点的影响，得到更好的决策边界，甚至当数据不是线性可分的时候，svm也可以得到好的结果。

运行程序：

%% Initialization
clear ; close all; clc

%% =============== Part 1: Loading and Visualizing Data ================
%  We start the exercise by first loading and visualizing the dataset.
%  The following code will load the dataset into your environment and plot
%  the data.
%

fprintf('Loading and Visualizing Data ...\n')

% Load from ex6data1:
% You will have X, y in your environment
load('ex6data1.mat');

% Plot training data
plotData(X, y);

fprintf('Program paused. Press enter to continue.\n');
pause;

%% ==================== Part 2: Training Linear SVM ====================
%  The following code will train a linear SVM on the dataset and plot the
%  decision boundary learned.
%

% Load from ex6data1:
% You will have X, y in your environment
load('ex6data1.mat');

fprintf('\nTraining Linear SVM ...\n')

% You should try to change the C value below and see how the decision
% boundary varies (e.g., try C = 1000)
C = 1;
model = svmTrain(X, y, C, @linearKernel, 1e-3, 20);
visualizeBoundaryLinear(X, y, model);

fprintf('Program paused. Press enter to continue.\n');
pause;

%% =============== Part 3: Implementing Gaussian Kernel ===============
%  You will now implement the Gaussian kernel to use
%  with the SVM. You should complete the code in gaussianKernel.m
%
fprintf('\nEvaluating the Gaussian Kernel ...\n')

x1 = [1 2 1]; x2 = [0 4 -1]; sigma = 2;
sim = gaussianKernel(x1, x2, sigma);

fprintf(['Gaussian Kernel between x1 = [1; 2; 1], x2 = [0; 4; -1], sigma = %f :' ...
         '\n\t%f\n(for sigma = 2, this value should be about 0.324652)\n'], sigma, sim);

fprintf('Program paused. Press enter to continue.\n');
pause;

%% =============== Part 4: Visualizing Dataset 2 ================
%  The following code will load the next dataset into your environment and
%  plot the data.
%

fprintf('Loading and Visualizing Data ...\n')

% Load from ex6data2:
% You will have X, y in your environment
load('ex6data2.mat');

% Plot training data
plotData(X, y);

fprintf('Program paused. Press enter to continue.\n');
pause;

%% ========== Part 5: Training SVM with RBF Kernel (Dataset 2) ==========
%  After you have implemented the kernel, we can now use it to train the
%  SVM classifier.
%
fprintf('\nTraining SVM with RBF Kernel (this may take 1 to 2 minutes) ...\n');

% Load from ex6data2:
% You will have X, y in your environment
load('ex6data2.mat');

% SVM Parameters
C = 1; sigma = 0.1;

% We set the tolerance and max_passes lower here so that the code will run
% faster. However, in practice, you will want to run the training to
% convergence.
model= svmTrain(X, y, C, @(x1, x2) gaussianKernel(x1, x2, sigma));
visualizeBoundary(X, y, model);

fprintf('Program paused. Press enter to continue.\n');
pause;

%% =============== Part 6: Visualizing Dataset 3 ================
%  The following code will load the next dataset into your environment and
%  plot the data.
%

fprintf('Loading and Visualizing Data ...\n')

% Load from ex6data3:
% You will have X, y in your environment
load('ex6data3.mat');

% Plot training data
plotData(X, y);

fprintf('Program paused. Press enter to continue.\n');
pause;

%% ========== Part 7: Training SVM with RBF Kernel (Dataset 3) ==========

%  This is a different dataset that you can use to experiment with. Try
%  different values of C and sigma here.
%

% Load from ex6data3:
% You will have X, y in your environment
load('ex6data3.mat');

% Try different SVM Parameters here
[C, sigma] = dataset3Params(X, y, Xval, yval);

% Train the SVM
model= svmTrain(X, y, C, @(x1, x2) gaussianKernel(x1, x2, sigma));
visualizeBoundary(X, y, model);

fprintf('Program paused. Press enter to continue.\n');
pause;