这周主要讲的是给定训练集下,为神经网络拟合参数的学习算法。
1、代价函数
由于在神经网络中可能会有多个输出结点,hΘ(x)k代表第k个输出的假设,所以神经网络的代价函数相当于是逻辑回归的代价函数的求和。
(没有正规化的公式,K表示输出结点的数量)
2、使用梯度下降求代价函数最小值,计算梯度就要求偏导,而BP反向传播法是一种有效的求偏导的方法。
(1)前向传播:
2、从后向前计算误差:
接下来看编程作业,由于我们只需要填写一些函数的空白的地方,所以为了弄懂整个流程,最好在命令窗口一行一行的敲ex4.m文件中的代码。
一、nnCostFunction.m:这个文件中要实现几个部分:
(1)前向传播,计算未正规化的代价函数J(θ);
a1 = X(i,:); %a1为X的行向量
a1 = a1'; %a1为列向量
% layer 2
z2 = Theta1 * a1; %theta1为25*401,a1为401*1, z2为25*1
a2 = sigmoid(z2); % a2为25*1列向量
a2 = [1; a2]; % 增加偏置单元1,a2为26*1
% layer 3
z3 = Theta2 * a2;
a3 = sigmoid(z3); % a3为输出,是一个列向量
p = zeros(num_labels, 1); % p是10*1
p(y(i)) = 1;
J = J + sum((-p).*log(a3) - (1-p).*log(1-a3));
(2)后向传播,计算更新后的参数
delta3 = a3 - p; % delta3为10*1
delta2 = Theta2(:,2:end)' * delta3 .* sigmoidGradient(z2); % delta2为25*1
Theta1_grad = Theta1_grad + delta2 * a1';
Theta2_grad = Theta2_grad + delta3 * a2';
然后求正规化的代价函数和后向传播,完整代码如下:
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
%
% 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.
%
% 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.
%
% Part1 前向传播,计算代价函数
X = [ones(m,1) X]; % 增加偏置单元1
for i = 1:m
a1 = X(i,:); %a1为X的行向量
a1 = a1'; %a1为列向量
% layer 2
z2 = Theta1 * a1; %theta1为25*401,a1为401*1, z2为25*1
a2 = sigmoid(z2); % a2为25*1列向量
a2 = [1; a2]; % 增加偏置单元1,a2为26*1
% layer 3
z3 = Theta2 * a2;
a3 = sigmoid(z3); % a3为输出,是一个列向量
p = zeros(num_labels, 1); % p是10*1
p(y(i)) = 1;
J = J + sum((-p).*log(a3) - (1-p).*log(1-a3));
% backpropagation;
delta3 = a3 - p; % delta3为10*1
delta2 = Theta2(:,2:end)' * delta3 .* sigmoidGradient(z2); % delta2为25*1
Theta1_grad = Theta1_grad + delta2 * a1';
Theta2_grad = Theta2_grad + delta3 * a2';
end
J = J / m;
Theta1_grad = Theta1_grad / m;
Theta2_grad = Theta2_grad / m;
% 2.正规化,不考虑偏置单元
temp1 = Theta1(:,2:size(Theta1,2)).^2;
temp2 = Theta2(:,2:size(Theta2,2)).^2;
reg = lambda / (2*m) * (sum(temp1(:)) + sum(temp2(:)));
J = J + reg;
% 3.反向传播正规化
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
二、randInitializeWeights.m:随机初始化神经网络的权重
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