ML - Coursera Andrew Ng - Week5 & Ex4 - Neural Network 2 - 笔记与代码

Week 5主要讲了神经网络的反向传播算法（Backpropagation Algorithm）。如上一节所讲，前馈传播算法是根据输入层向量x，各层权重矩阵$\Theta$ ，以及激活函数逐层从左向右计算，最终得到输出层结果。反向传播算法则是为了找出使损失函数$J(\Theta )$最小化的最佳$\Theta$ ，BP为优化函数（如梯度下降、其它高级优化算法）提供梯度值，通过从右向左来逐层求偏导，得到每层的参数梯度，用${\delta }^{(l)}$来保存每一层求偏导的部分结果，以便后面可以接着借用这部分结果来求下一层的结果，用$D_{i,j}^{(l)}$来保存偏导结果。

选择神经网络的结构。

训练神经网络的时候，需要先随机化权重$\Theta$，不能初始化为0，不然BP的时候所有节点会重复地更新为同一个值。还要注意用gradient checking来保证梯度下降算法或其它高级算法的正确性，因为算法的复杂性，有时候一些bug会导致表面上梯度是下降的，但是结果会有很大误差。

Cost Function

正则化逻辑回归的损失函数为

神经网络的损失函数为

其中L为神经网络的总层数，$s_l$为第l层不包括bias unit的单元数量，K为输出层的units/classes数量。

• 前一项的两个求和表示计算输出层每个单元的逻辑回归损失之和
• 后一项的三个求和表示计算整个神经网络中的$\Theta$的各项平方之和。在第l层中$s_{l+1}$是当前$\Theta$的行数（不包括bias unit）， $s_l$是当前$\Theta$的列数（包括bias unit）。

Backpropagation Algorithm

反向传播算法是为了得到
$$mi{n}_{\Theta }J(\Theta )$$

所以我们用损失函数对$\Theta$求偏导（partial derivative）
$$\frac{\partial }{\partial {\Theta }_{i,j}^{(l)}}J(\Theta )$$

考虑只有一个分类（K=1）的简单情况，$J(\Theta )$可以简化为
$$J(\Theta )=cost(t)=y^{(t)}log(h_{\Theta }(x^{(t)}))+(1-y^{(t)})log(1-h_{\Theta }(x^{(t)}))$$

${\delta }_j^{(l)}$可以看做是$a_j^{(l)}$处的误差"error"，具体推导可见link，定义${\delta }^{(l)}$为
$${\delta }_j^{(l)}=\frac{\partial }{\partial z_j^{(l)}}J(\Theta )$$

${\delta }^{(l)}$的大小与对应那一层的$a^{(l)}$的大小一样，${\Delta }^{(l)}$的大小与对应那一层的${\Theta }^{(l)}$的大小一样，通过${\Delta }^{(l)}$来更新${\Theta }^{(l)}$，找到使损失函数最小的$\Theta$。BP算法的计算过程：

训练集{($x^{(1)}$, $y^{(1)}$)…($x^{(m)}$, $y^{(m)}$)}，初始化${\Delta }_{i,j}^{(l)}:=0$
对训练样本t = 1 to m:

1. Set $a^{(1)}:=x^{(t)}$
2. 使用前馈算法计算后面每层的结果$a^{(l)}$
3. 计算${\delta }^{(l)}$:
   ${\delta }^{(L)}=a^{(L)}-y^{(t)}$，$l=L$
   ${\delta }^{(l)}=(({\Theta }^{(l)}{)}^T{\delta }^{(l+1)}.\ast a^{(l)}.\ast (1-a^{(l)})$，$l=L-1,L-2,...,2$
4. 计算${\Delta }_{i,j}^{(l)}$：
   ${\Delta }_{i,j}^{(l)}:={\Delta }_{i,j}^{(l)}+a_j^{(l)}{\delta }^{(l+1)}$
   vectorization ${\Delta }^{(l)}:={\Delta }^{(l)}+{\delta }^{(l+1)}(a_j^{(l)}{)}^T$
5. 计算$D_{i,j}^{(l)}$:
   $D_{i,j}^{(l)}:=\frac{1}{m}({\Delta }_{i,j}^{(l)}+\lambda {\Theta }_{i,j}^{(l)})$，$j̸=0$
   $D_{i,j}^{(l)}:=\frac{1}{m}{\Delta }_{i,j}^{(l)}$，$j=0$

Unrolling Parameters

为了能使用优化函数，如fminunc()，需要将unroll所有元素将它们放进一个vector中。如Theta1是个25x401的矩阵，Theta2是个10x26的矩阵，unroll后是一个10285x1的vector。

thetaVector = [ Theta1(:); Theta2(:); Theta3(:); ]
deltaVector = [ D1(:); D2(:); D3(:) ]

// if the dimensions of Theta1 is 10x11, Theta2 is 10x11 and Theta3 is 1x11
Theta1 = reshape(thetaVector(1:110),10,11)
Theta2 = reshape(thetaVector(111:220),10,11)
Theta3 = reshape(thetaVector(221:231),1,11)

Gradient Checking

Gradient checking可以保证算法的正确性，通过检查gradApprox ≈ deltaVector来确认算法是否正确。一旦确认BP算法的正确性，就不需要继续计算gradApprox了，因为gradient checking远比BP耗时。

epsilon = 1e-4;
for i = 1:n,
  thetaPlus = theta;
  thetaPlus(i) += epsilon;
  thetaMinus = theta;
  thetaMinus(i) -= epsilon;
  gradApprox(i) = (J(thetaPlus) - J(thetaMinus))/(2*epsilon)
end;

Random Initialization

将每个${\Theta }_{i,j}^{(l)}$初始化为$[-ϵ,ϵ]$之间的随机值，此处代码中的epsilon与gradient checking中的epsilon无关。

// If the dimensions of Theta1 is 10x11, Theta2 is 10x11 and Theta3 is 1x11.
// rand(x,y) is just a function in octave that will initialize a matrix of random real numbers between 0 and 1.
Theta1 = rand(10,11) * (2 * INIT_EPSILON) - INIT_EPSILON;
Theta2 = rand(10,11) * (2 * INIT_EPSILON) - INIT_EPSILON;
Theta3 = rand(1,11) * (2 * INIT_EPSILON) - INIT_EPSILON;

Exercise 4: 实现反向传播的神经网络 - Matlab

用反向传播算法实现神经网络，来识别手写数字。

1. Neural Networks

构建的神经网络结构与ex3中一致，3层神经网络，输入层特征数量为400（不包括bias unit），输出层类别数量为10，隐藏层单元数为25。注意unroll parameters。

1.1 Feedforward and Cost Function

非正则化的神经网络损失函数为

在nnCostFunction.m文件中，实现非正则化的损失函数。因为没有正则化，lambda设为0。对ex3中给出的Theta1和Theta2，我们可以测试这个损失函数的正确性。eye(num_labels)(:,y)的解释参考link。sum(x)表示列求和，sum(x,2)表示行求和，sum(x($:$))表示矩阵求和。

% feedforward propagation
X = [ones(m, 1), X]; % 5000x401
a_super_2 = sigmoid(Theta1 * X'); % 25x5000=(25x401)*(401x5000)
a_super_2 = [ones(1, m); a_super_2]; % 26x5000
a_super_3 = sigmoid(Theta2 * a_super_2); % 10x5000=(10x26)*(26*5000)

% non-regularized cost function
I = eye(num_labels); % 10x10
y = I(:,y); % y:5000x1 -> 10x5000
tempJ = (y .* log(a_super_3) - (y - 1) .* log(1 - a_super_3)) / (-m);
J = sum(tempJ(:));

1.2 Regularized Cost Function

正则化的神经网络损失函数为

在nnCostFunction.m文件中，基于上面非正则化的损失函数基础上，加上正则化的项，即可得到正则化的损失函数。

% regularized cost function
tempTheta1 = Theta1.^2;
tempTheta1 = tempTheta1(:,2:end);
tempTheta2 = Theta2.^2;
tempTheta2 = tempTheta2(:,2:end);
J = J + lambda / (2 * m) * (sum(tempTheta1(:)) + sum(tempTheta2(:)));

2. Backpropagation

2.1 Sigmoid Gradient

sigmoid函数求导有如下特征，在sigmoidGradient.m文件中完成对sigmoid的求导。

function g = sigmoidGradient(z)
%SIGMOIDGRADIENT returns the gradient of the sigmoid function
%evaluated at z
%   g = SIGMOIDGRADIENT(z) computes the gradient of the sigmoid function
%   evaluated at z. This should work regardless if z is a matrix or a
%   vector. In particular, if z is a vector or matrix, you should return
%   the gradient for each element.

g = zeros(size(z));

% ====================== YOUR CODE HERE ======================
% Instructions: Compute the gradient of the sigmoid function evaluated at
%               each value of z (z can be a matrix, vector or scalar).

g = sigmoid(z) .* (1 - sigmoid(z));

% =============================================================

end

2.2 Random Initialization

在randInitializeWeights.m文件中，完成对参数的随机初始化。

function W = randInitializeWeights(L_in, L_out)
%RANDINITIALIZEWEIGHTS Randomly initialize the weights of a layer with L_in
%incoming connections and L_out outgoing connections
%   W = RANDINITIALIZEWEIGHTS(L_in, L_out) randomly initializes the weights 
%   of a layer with L_in incoming connections and L_out outgoing 
%   connections. 
%
%   Note that W should be set to a matrix of size(L_out, 1 + L_in) as
%   the first column of W handles the "bias" terms
%

% You need to return the following variables correctly 
W = zeros(L_out, 1 + L_in);

% ====================== YOUR CODE HERE ======================
% Instructions: Initialize W randomly so that we break the symmetry while
%               training the neural network.
%
% Note: The first column of W corresponds to the parameters for the bias unit
%

epsilon_init = 0.12;
W = rand(L_out, 1 + L_in) * 2 * epsilon_init - epsilon_init;

% =========================================================================

end

2.3 Backpropagation

BP需要从右到左计算输出层和隐藏层的"error term" ${\delta }_j^{(l)}$，计算方式如下图所示。一共有m=5000个样本，对每个样本，都要进行1-4的操作，最后将得到的gradients的和除以m就是整个神经网络损失函数的gradients。${\delta }^{(l)}$的大小与对应那一层的$a^{(l)}$的大小一样，${\Delta }^{(l)}$的大小与对应那一层的${\Theta }^{(l)}$的大小一样。

delta_super_3 = a_super_3 - y; % 10x5000
delta_super_2 = Theta2(:,2:end)' * delta_super_3 .* sigmoidGradient(Theta1 * X'); % 25x5000

Delta_super_2 = delta_super_3 * a_super_2'; % 10x26=(10x5000)*(5000x26)
Delta_super_1 = delta_super_2 * X; % 25x401=(25x5000)*(5000x401)

% gradient for non-regularized BP
Theta1_grad = Delta_super_1 / m;
Theta2_grad = Delta_super_2 / m;

2.4 Gradient checking

Gradient checking可以用于各种算法的检验。此处不需要自己填写代码，比较numerical gradient和我们得到的gradient是否在一个很小的误差内，此处的$ϵ=10^{-4}$。在确认了算法正确性后，训练参数前记得关闭gradient checking。

2.5 Regularized Neural Network

在2.3的基础上，给gradient增加上正则化项，就可以得到正则化的gradient，注意不能正则化${\Theta }^{(l)}$的第一列，因为那部分是用于bias unit的，也就是说${\Theta }_{i,j}^{(l)}$中，我们需要用到的部分i从1开始，j从0开始。在nnCostFunction.m文件中加上正则化后的gradients。

% regularized gradients
Theta1(:,1) = 0;
Theta2(:,1) = 0;
Theta1_grad = Theta1_grad + lambda / m * Theta1;
Theta2_grad = Theta2_grad + lambda / m * Theta2;

在nnCostFunction.m文件中，我们间断地补充了几次代码，主要进行了三步操作，完整的代码如下。

function [J grad] = nnCostFunction(nn_params, ...
                                   input_layer_size, ...
                                   hidden_layer_size, ...
                                   num_labels, ...
                                   X, y, lambda)
%NNCOSTFUNCTION Implements the neural network cost function for a two layer
%neural network which performs classification
%   [J grad] = NNCOSTFUNCTON(nn_params, hidden_layer_size, num_labels, ...
%   X, y, lambda) computes the cost and gradient of the neural network. The
%   parameters for the neural network are "unrolled" into the vector
%   nn_params and need to be converted back into the weight matrices. 
% 
%   The returned parameter grad should be a "unrolled" vector of the
%   partial derivatives of the neural network.
%

% Reshape nn_params back into the parameters Theta1 and Theta2, the weight matrices
% for our 2 layer neural network
Theta1 = reshape(nn_params(1:hidden_layer_size * (input_layer_size + 1)), ...
                 hidden_layer_size, (input_layer_size + 1));

Theta2 = reshape(nn_params((1 + (hidden_layer_size * (input_layer_size + 1))):end), ...
                 num_labels, (hidden_layer_size + 1));

% Setup some useful variables
m = size(X, 1);
         
% You need to return the following variables correctly 
J = 0;
Theta1_grad = zeros(size(Theta1));
Theta2_grad = zeros(size(Theta2));

% ====================== YOUR CODE HERE ======================
% Instructions: You should complete the code by working through the
%               following parts.
%
% Part 1: Feedforward the neural network and return the cost in the
%         variable J. After implementing Part 1, you can verify that your
%         cost function computation is correct by verifying the cost
%         computed in ex4.m

% feedforward propagation
X = [ones(m, 1), X]; % 5000x401
a_super_2 = sigmoid(Theta1 * X'); % 25x5000=(25x401)*(401x5000)
a_super_2 = [ones(1, m); a_super_2]; % 26x5000
a_super_3 = sigmoid(Theta2 * a_super_2); % 10x5000=(10x26)*(26*5000)

% non-regularized cost function
I = eye(num_labels); % 10x10
y = I(:,y); % y:5000x1 -> 10x5000
tempJ = (y .* log(a_super_3) - (y - 1) .* log(1 - a_super_3)) / (-m);
J = sum(tempJ(:));

% regularized cost function
tempTheta1 = Theta1.^2;
tempTheta1 = tempTheta1(:,2:end);
tempTheta2 = Theta2.^2;
tempTheta2 = tempTheta2(:,2:end);
J = J + lambda / (2 * m) * (sum(tempTheta1(:)) + sum(tempTheta2(:)));

% Part 2: Implement the backpropagation algorithm to compute the gradients
%         Theta1_grad and Theta2_grad. You should return the partial derivatives of
%         the cost function with respect to Theta1 and Theta2 in Theta1_grad and
%         Theta2_grad, respectively. After implementing Part 2, you can check
%         that your implementation is correct by running checkNNGradients
%
%         Note: The vector y passed into the function is a vector of labels
%               containing values from 1..K. You need to map this vector into a 
%               binary vector of 1's and 0's to be used with the neural network
%               cost function.
%
%         Hint: We recommend implementing backpropagation using a for-loop
%               over the training examples if you are implementing it for the 
%               first time.

delta_super_3 = a_super_3 - y; % 10x5000
delta_super_2 = Theta2(:,2:end)' * delta_super_3 .* sigmoidGradient(Theta1 * X'); % 25x5000

Delta_super_2 = delta_super_3 * a_super_2'; % 10x26=(10x5000)*(5000x26)
Delta_super_1 = delta_super_2 * X; % 25x401=(25x5000)*(5000x401)

% gradient for non-regularized BP
Theta1_grad = Delta_super_1 / m;
Theta2_grad = Delta_super_2 / m;

% Part 3: Implement regularization with the cost function and gradients.
%
%         Hint: You can implement this around the code for
%               backpropagation. That is, you can compute the gradients for
%               the regularization separately and then add them to Theta1_grad
%               and Theta2_grad from Part 2.
%

% regularized gradients
Theta1(:,1) = 0;
Theta2(:,1) = 0;
Theta1_grad = Theta1_grad + lambda / m * Theta1;
Theta2_grad = Theta2_grad + lambda / m * Theta2;

% -------------------------------------------------------------

% =========================================================================

% Unroll gradients
grad = [Theta1_grad(:) ; Theta2_grad(:)];

end

2.6 Learning Parameters Using fmincg

用fmincg训练参数，注意参数是unroll了的。

options = optimset('MaxIter', 50);
lambda = 1;

% Create "short hand" for the cost function to be minimized
costFunction = @(p) nnCostFunction(p, input_layer_size, hidden_layer_size, num_labels, X, y, lambda);

% Now, costFunction is a function that takes in only one argument (the
% neural network parameters)
[nn_params, ~] = fmincg(costFunction, initial_nn_params, options);

% Obtain Theta1 and Theta2 back from nn_params
Theta1 = reshape(nn_params(1:hidden_layer_size * (input_layer_size + 1)), hidden_layer_size, (input_layer_size + 1));
Theta2 = reshape(nn_params((1 + (hidden_layer_size * (input_layer_size + 1))):end), num_labels, (hidden_layer_size + 1));
 
pred = predict(Theta1, Theta2, X);
fprintf('\nTraining Set Accuracy: %f\n', mean(double(pred == y)) * 100);

3. Visualizing the Hidden Layer

我们可以通过可视化隐藏层来看神经网络学习的是什么东西。

Ex4全部代码已上传Github。