Sparse Autoencoder Exercise

本文通过用sparse autoencoder 方法训练输入的图片集，从而得出一个隐藏层所训练的特征。理论文档参考：http://deeplearning.stanford.edu/wiki/index.php/Neural_Networks实验文档参考：http://deeplearning.stanford.edu/wiki/index.php/Exercise:Sparse_Autoencoder。训练样本为10000张8*8 大小的图片，网络总共3层，输入层有8*8个节点，隐藏层有25个节点，输出为8*8个节点。通过训练，使输出等于输入。

实验步骤

1、生成训练数据集

通过对512*512大小的10幅图像随机采样10000张8*8的图片，然后将每一张图片重新排列成一列，形成64*10000的训练数据集。

2、实现稀疏自编码器

这一步计算代价函数和相应的偏导数（采用反向传导），详细说明请看理论文档，这里仅给出计算公式：

代价函数：

对应的偏导数通过反向传导求出，首先求出输出层的残差值，即该层输入加权值对该层输出与理想输出之差的影响。

对于其他网络层，有

由于加入了稀疏性限制，对于隐藏层，有：

那么，代价函数对各层权重和偏置量的偏导如下：

上述公式中m 表示训练样本数，λ表示正规化参数，ρ 表示稀疏性参数，β 表示稀疏限制权重因子，其他字母含义见理论文档。

3、梯度值检验

通过一个给定的示例验证所写的梯度求解方法的正确性。

4、训练自编码器

通过L-BFGS算法优化代价函数。

5、可视化学习到的权重值

实验代码

实验数据包里面已将大致的实验流程代码写出，见train.m。

train.m:

%% CS294A/CS294W Programming Assignment Starter Code

%  Instructions
%  ------------
% 
%  This file contains code that helps you get started on the
%  programming assignment. You will need to complete the code in sampleIMAGES.m,
%  sparseAutoencoderCost.m and computeNumericalGradient.m. 
%  For the purpose of completing the assignment, you do not need to
%  change the code in this file. 
%
%%======================================================================
%% STEP 0: Here we provide the relevant parameters values that will
%  allow your sparse autoencoder to get good filters; you do not need to 
%  change the parameters below.

visibleSize = 8*8;   % number of input units 
hiddenSize = 25;     % number of hidden units 
sparsityParam = 0.01;   % desired average activation of the hidden units.
                     % (This was denoted by the Greek alphabet rho, which looks like a lower-case "p",
		     %  in the lecture notes). 
lambda = 0.0001;     % weight decay parameter       
beta = 3;            % weight of sparsity penalty term       

%%======================================================================
%% STEP 1: Implement sampleIMAGES
%
%  After implementing sampleIMAGES, the display_network command should
%  display a random sample of 200 patches from the dataset

patches = sampleIMAGES;
display_network(patches(:,randi(size(patches,2),200,1)),8);


%  Obtain random parameters theta
theta = initializeParameters(hiddenSize, visibleSize);

%%======================================================================
%% STEP 2: Implement sparseAutoencoderCost
%
%  You can implement all of the components (squared error cost, weight decay term,
%  sparsity penalty) in the cost function at once, but it may be easier to do 
%  it step-by-step and run gradient checking (see STEP 3) after each step.  We 
%  suggest implementing the sparseAutoencoderCost function using the following steps:
%
%  (a) Implement forward propagation in your neural network, and implement the 
%      squared error term of the cost function.  Implement backpropagation to 
%      compute the derivatives.   Then (using lambda=beta=0), run Gradient Checking 
%      to verify that the calculations corresponding to the squared error cost 
%      term are correct.
%
%  (b) Add in the weight decay term (in both the cost function and the derivative
%      calculations), then re-run Gradient Checking to verify correctness. 
%
%  (c) Add in the sparsity penalty term, then re-run Gradient Checking to 
%      verify correctness.
%
%  Feel free to change the training settings when debugging your
%  code.  (For example, reducing the training set size or 
%  number of hidden units may make your code run faster; and setting beta 
%  and/or lambda to zero may be helpful for debugging.)  However, in your 
%  final submission of the visualized weights, please use parameters we 
%  gave in Step 0 above.

[cost, grad] = sparseAutoencoderCost(theta, visibleSize, hiddenSize, lambda, ...
                                     sparsityParam, beta, patches);

%%======================================================================
%% STEP 3: Gradient Checking
%
% Hint: If you are debugging your code, performing gradient checking on smaller models 
% and smaller training sets (e.g., using only 10 training examples and 1-2 hidden 
% units) may speed things up.

% First, lets make sure your numerical gradient computation is correct for a
% simple function.  After you have implemented computeNumericalGradient.m,
% run the following: 
checkNumericalGradient();

% Now we can use it to check your cost function and derivative calculations
% for the sparse autoencoder.  
numgrad = computeNumericalGradient( @(x) sparseAutoencoderCost(x, visibleSize, ...
                                                  hiddenSize, lambda, ...
                                                  sparsityParam, beta, ...
                                                  patches), theta);

% Use this to visually compare the gradients side by side
disp([numgrad grad]); 

% Compare numerically computed gradients with the ones obtained from backpropagation
diff = norm(numgrad-grad)/norm(numgrad+grad);
disp(diff); % Should be small. In our implementation, these values are
            % usually less than 1e-9.

            % When you got this working, Congratulations!!! 

%%======================================================================
%% STEP 4: After verifying that your implementation of
%  sparseAutoencoderCost is correct, You can start training your sparse
%  autoencoder with minFunc (L-BFGS).

%  Randomly initialize the parameters
theta = initializeParameters(hiddenSize, visibleSize);

%  Use minFunc to minimize the function
addpath minFunc/
options.Method = 'lbfgs'; % Here, we use L-BFGS to optimize our cost
                          % function. Generally, for minFunc to work, you
                          % need a function pointer with two outputs: the
                          % function value and the gradient. In our problem,
                          % sparseAutoencoderCost.m satisfies this.
options.maxIter = 400;	  % Maximum number of iterations of L-BFGS to run 
options.display = 'on';


[opttheta, cost] = minFunc( @(p) sparseAutoencoderCost(p, ...
                                   visibleSize, hiddenSize, ...
                                   lambda, sparsityParam, ...
                                   beta, patches), ...
                              theta, options);

%%======================================================================
%% STEP 5: Visualization 

W1 = reshape(opttheta(1:hiddenSize*visibleSize), hiddenSize, visibleSize);
display_network(W1', 12);

print -djpeg weights.jpg   % save the visualization to a file 

第一步的采样训练数据集（这里每一幅图像我都采了(numpatches / k) 次，没有随机。

sampleIMAGES.m

function patches = sampleIMAGES()
% sampleIMAGES
% Returns 10000 patches for training

load IMAGES;    % load images from disk 

patchsize = 8;  % we'll use 8x8 patches 
numpatches = 10000;

% Initialize patches with zeros.  Your code will fill in this matrix--one
% column per patch, 10000 columns. 
patches = zeros(patchsize*patchsize, numpatches);

%% ---------- YOUR CODE HERE --------------------------------------
%  Instructions: Fill in the variable called "patches" using data 
%  from IMAGES.  
%  
%  IMAGES is a 3D array containing 10 images
%  For instance, IMAGES(:,:,6) is a 512x512 array containing the 6th image,
%  and you can type "imagesc(IMAGES(:,:,6)), colormap gray;" to visualize
%  it. (The contrast on these images look a bit off because they have
%  been preprocessed using using "whitening."  See the lecture notes for
%  more details.) As a second example, IMAGES(21:30,21:30,1) is an image
%  patch corresponding to the pixels in the block (21,21) to (30,30) of
%  Image 1

%  IMAGES should be graysets
[m, n, k] = size(IMAGES);
for i = 1 : numpatches
    index = ceil([m - patchsize + 1; n - patchsize + 1] .* rand(2, 1));
    imagePatch = IMAGES(index(1) : index(1) + patchsize - 1,...
        index(2) : index(2) + patchsize - 1, mod(i, 10) + 1);
    patches(:, i) = imagePatch(:);
end

%% ---------------------------------------------------------------
% For the autoencoder to work well we need to normalize the data
% Specifically, since the output of the network is bounded between [0,1]
% (due to the sigmoid activation function), we have to make sure 
% the range of pixel values is also bounded between [0,1]
patches = normalizeData(patches);

end


%% ---------------------------------------------------------------
function patches = normalizeData(patches)

% Squash data to [0.1, 0.9] since we use sigmoid as the activation
% function in the output layer

% Remove DC (mean of images). 
patches = bsxfun(@minus, patches, mean(patches));

% Truncate to +/-3 standard deviations and scale to -1 to 1
pstd = 3 * std(patches(:));
patches = max(min(patches, pstd), -pstd) / pstd;

% Rescale from [-1,1] to [0.1,0.9]
patches = (patches + 1) * 0.4 + 0.1;

end

第二步的代价函数及偏导数的实现。

sparseAutoencoderCost.m:

function [cost,grad] = sparseAutoencoderCost(theta, visibleSize, hiddenSize, ...
                                             lambda, sparsityParam, beta, data)

% visibleSize: the number of input units (probably 64) 
% hiddenSize: the number of hidden units (probably 25) 
% lambda: weight decay parameter
% sparsityParam: The desired average activation for the hidden units (denoted in the lecture
%                           notes by the greek alphabet rho, which looks like a lower-case "p").
% beta: weight of sparsity penalty term
% data: Our 64x10000 matrix containing the training data.  So, data(:,i) is the i-th training example. 
  
% The input theta is a vector (because minFunc expects the parameters to be a vector). 
% We first convert theta to the (W1, W2, b1, b2) matrix/vector format, so that this 
% follows the notation convention of the lecture notes. 

W1 = reshape(theta(1:hiddenSize*visibleSize), hiddenSize, visibleSize);
W2 = reshape(theta(hiddenSize*visibleSize+1:2*hiddenSize*visibleSize), visibleSize, hiddenSize);
b1 = theta(2*hiddenSize*visibleSize+1:2*hiddenSize*visibleSize+hiddenSize);
b2 = theta(2*hiddenSize*visibleSize+hiddenSize+1:end);

% Cost and gradient variables (your code needs to compute these values). 
% Here, we initialize them to zeros. 
cost = 0;
W1grad = zeros(size(W1)); 
W2grad = zeros(size(W2));
b1grad = zeros(size(b1)); 
b2grad = zeros(size(b2));

%% ---------- YOUR CODE HERE --------------------------------------
%  Instructions: Compute the cost/optimization objective J_sparse(W,b) for the Sparse Autoencoder,
%                and the corresponding gradients W1grad, W2grad, b1grad, b2grad.
%
% W1grad, W2grad, b1grad and b2grad should be computed using backpropagation.
% Note that W1grad has the same dimensions as W1, b1grad has the same dimensions
% as b1, etc.  Your code should set W1grad to be the partial derivative of J_sparse(W,b) with
% respect to W1.  I.e., W1grad(i,j) should be the partial derivative of J_sparse(W,b) 
% with respect to the input parameter W1(i,j).  Thus, W1grad should be equal to the term 
% [(1/m) \Delta W^{(1)} + \lambda W^{(1)}] in the last block of pseudo-code in Section 2.2 
% of the lecture notes (and similarly for W2grad, b1grad, b2grad).
% 
% Stated differently, if we were using batch gradient descent to optimize the parameters,
% the gradient descent update to W1 would be W1 := W1 - alpha * W1grad, and similarly for W2, b1, b2. 
% 

rho = zeros(hiddenSize, 1); % average activation value of hidden layer

m = size(data, 2);
% forward propagation
for i = 1 : m
    Z2 = W1 * data(:, i) + b1;
    A2 = sigmoid(Z2);
    rho = rho + A2;
end
rho = (1 / m) * rho;

% backward propagation
for i = 1 : m
    Z2 = W1 * data(:, i) + b1;  % weighted input value
    A2 = sigmoid(Z2);           % activaton value
    Z3 = W2 * A2 + b2;
    A3 = sigmoid(Z3);
    
    cost = cost + 1 / (2 * m) * (A3 - data(:, i))' * (A3 - data(:, i)); %mean square error

    % compute error term
    % For output layer nl
    delta3 = (A3 - data(:, i)) .* (A3 .* (1 - A3));
    % For l = nl -1, nl -2...2 layer
    delta2 = (W2' * delta3 + beta * ...
        (-sparsityParam ./ rho + (1 - sparsityParam) ./ (1 - rho))) .* (A2 .* (1 - A2));
    
    % gradient
    W2grad = W2grad + delta3 * A2';
    W1grad = W1grad + delta2 * data(:, i)';
    b2grad = b2grad + delta3;
    b1grad = b1grad + delta2;
end

KL_divergence = sparsityParam * log(sparsityParam ./ rho) + ...
    (1 - sparsityParam) * log((1 - sparsityParam) ./ (1 - rho));
cost = cost + lambda / 2 * (sum(sum(W1.^2)) + sum(sum(W2.^2))) + ...
    beta * sum(KL_divergence);  % cost = mean square error + regularization + sparsity

W1grad = (1 / m) * W1grad + lambda * W1;
W2grad = (1 / m) * W2grad + lambda * W2;
b1grad = (1 / m) * b1grad;
b2grad = (1 / m) * b2grad;

%-------------------------------------------------------------------
% After computing the cost and gradient, we will convert the gradients back
% to a vector format (suitable for minFunc).  Specifically, we will unroll
% your gradient matrices into a vector.

grad = [W1grad(:) ; W2grad(:) ; b1grad(:) ; b2grad(:)];

end

%-------------------------------------------------------------------
% Here's an implementation of the sigmoid function, which you may find useful
% in your computation of the costs and the gradients.  This inputs a (row or
% column) vector (say (z1, z2, z3)) and returns (f(z1), f(z2), f(z3)). 

function sigm = sigmoid(x)
  
    sigm = 1 ./ (1 + exp(-x));
end

第三步计算函数梯度值。

computeNumericalGradient.m:

function numgrad = computeNumericalGradient(J, theta)
% numgrad = computeNumericalGradient(J, theta)
% theta: a vector of parameters
% J: a function that outputs a real-number. Calling y = J(theta) will return the
% function value at theta. 
  
% Initialize numgrad with zeros
numgrad = zeros(size(theta));

%% ---------- YOUR CODE HERE --------------------------------------
% Instructions: 
% Implement numerical gradient checking, and return the result in numgrad.  
% (See Section 2.3 of the lecture notes.)
% You should write code so that numgrad(i) is (the numerical approximation to) the 
% partial derivative of J with respect to the i-th input argument, evaluated at theta.  
% I.e., numgrad(i) should be the (approximately) the partial derivative of J with 
% respect to theta(i).
%                
% Hint: You will probably want to compute the elements of numgrad one at a time. 

EPSILON = 0.0001;
m = length(theta);
for i = 1 : m
    e = zeros(m, 1);
    e(i) = 1;
    numgrad(i) = (J(theta + EPSILON * e) - J(theta - EPSILON * e)) / (2 * EPSILON); 
end

%% ---------------------------------------------------------------
end

程序运行时间较长，大约一个半小时，主要是梯度计算耗时，计算所得的梯度误差在7.2591e-11，小于1e-9，说明代价函数和偏导数求解正确。其训练所得隐藏层特征如下：