Day 4

Data Analysis and Machine Learning:
Neural networks, from the simple

perceptron to deep learning
Morten Hjorth-Jensen1,2
1
Department of Physics and Center for Computing in Science Education, University of Oslo, Norway
partment of Physics and Astronomy and Facility for Rare Ion Beams and National Superconducting Cyclotron Laboratory, Michigan State University, U
Jun 26, 2020
Neural networks
Artificial neural networks are computational systems that can learn to perform
tasks by considering examples, generally without being programmed with any
task-specific rules. It is supposed to mimic a biological system, wherein neurons
interact by sending signals in the form of mathematical functions between layers.
All layers can contain an arbitrary number of neurons, and each connection is
represented by a weight variable.
Artificial neurons
The field of artificial neural networks has a long history of development, and
is closely connected with the advancement of computer science and computers
in general. A model of artificial neurons was first developed by McCulloch and
Pitts in 1943 to study signal processing in the brain and has later been refined
by others. The general idea is to mimic neural networks in the human brain,
which is composed of billions of neurons that communicate with each other by
sending electrical signals. Each neuron accumulates its incoming signals, which
must exceed an activation threshold to yield an output. If the threshold is not
overcome, the neuron remains inactive, i.e. has zero output.
This behaviour has inspired a simple mathematical model for an artificial
neuron.
n
!
X
y=f wi xi = f (u) (1)
i=1
c 1999-2020, Morten Hjorth-Jensen. Released under CC Attribution-NonCommercial 4.0

license
Here, the output y of the neuron is the value of its activation function, which
have as input a weighted sum of signals xi , . . . , xn received by n other neurons.
Conceptually, it is helpful to divide neural networks into four categories:
1. general purpose neural networks for supervised learning,
2. neural networks designed specifically for image processing, the most promi-
nent example of this class being Convolutional Neural Networks (CNNs),
3. neural networks for sequential data such as Recurrent Neural Networks
(RNNs), and
4. neural networks for unsupervised learning such as Deep Boltzmann Ma-
chines.
In natural science, DNNs and CNNs have already found numerous applications.
In statistical physics, they have been applied to detect phase transitions in 2D
Ising and Potts models, lattice gauge theories, and different phases of polymers,
or solving the Navier-Stokes equation in weather forecasting. Deep learning has
also found interesting applications in quantum physics. Various quantum phase
transitions can be detected and studied using DNNs and CNNs, topological
phases, and even non-equilibrium many-body localization. Representing quantum
states as DNNs quantum state tomography are among some of the impressive
achievements to reveal the potential of DNNs to facilitate the study of quantum
systems.
In quantum information theory, it has been shown that one can perform gate
decompositions with the help of neural.
The applications are not limited to the natural sciences. There is a plethora
of applications in essentially all disciplines, from the humanities to life science
and medicine.
Neural network types

An artificial neural network (ANN), is a computational model that consists of lay-
ers of connected neurons, or nodes or units. We will refer to these interchangeably
as units or nodes, and sometimes as neurons.
It is supposed to mimic a biological nervous system by letting each neuron
interact with other neurons by sending signals in the form of mathematical
functions between layers. A wide variety of different ANNs have been developed,
but most of them consist of an input layer, an output layer and eventual layers
in-between, called hidden layers. All layers can contain an arbitrary number
of nodes, and each connection between two nodes is associated with a weight
variable.
Neural networks (also called neural nets) are neural-inspired nonlinear models
for supervised learning. As we will see, neural nets can be viewed as natural,
more powerful extensions of supervised learning methods such as linear and
logistic regression and soft-max methods we discussed earlier.
2
Feed-forward neural networks
The feed-forward neural network (FFNN) was the first and simplest type of
ANNs that were devised. In this network, the information moves in only one
direction: forward through the layers.
Nodes are represented by circles, while the arrows display the connections
between the nodes, including the direction of information flow. Additionally,
each arrow corresponds to a weight variable (figure to come). We observe that
each node in a layer is connected to all nodes in the subsequent layer, making
this a so-called fully-connected FFNN.
Convolutional Neural Network

A different variant of FFNNs are convolutional neural networks (CNNs), which
have a connectivity pattern inspired by the animal visual cortex. Individual
neurons in the visual cortex only respond to stimuli from small sub-regions of
the visual field, called a receptive field. This makes the neurons well-suited to
exploit the strong spatially local correlation present in natural images. The
response of each neuron can be approximated mathematically as a convolution
operation. (figure to come)
Convolutional neural networks emulate the behaviour of neurons in the visual
cortex by enforcing a local connectivity pattern between nodes of adjacent layers:
Each node in a convolutional layer is connected only to a subset of the nodes
in the previous layer, in contrast to the fully-connected FFNN. Often, CNNs
consist of several convolutional layers that learn local features of the input, with
a fully-connected layer at the end, which gathers all the local data and produces
the outputs. They have wide applications in image and video recognition.
Recurrent neural networks

So far we have only mentioned ANNs where information flows in one direction:
forward. Recurrent neural networks on the other hand, have connections between
nodes that form directed cycles. This creates a form of internal memory which
are able to capture information on what has been calculated before; the output is
dependent on the previous computations. Recurrent NNs make use of sequential
information by performing the same task for every element in a sequence, where
each element depends on previous elements. An example of such information
is sentences, making recurrent NNs especially well-suited for handwriting and
speech recognition.
Other types of networks

There are many other kinds of ANNs that have been developed. One type that
is specifically designed for interpolation in multidimensional space is the radial
basis function (RBF) network. RBFs are typically made up of three layers: an
input layer, a hidden layer with non-linear radial symmetric activation functions
3
and a linear output layer (”linear” here means that each node in the output layer
has a linear activation function). The layers are normally fully-connected and
there are no cycles, thus RBFs can be viewed as a type of fully-connected FFNN.
They are however usually treated as a separate type of NN due the unusual
activation functions.
Multilayer perceptrons
One uses often so-called fully-connected feed-forward neural networks with three
or more layers (an input layer, one or more hidden layers and an output layer)
consisting of neurons that have non-linear activation functions.
Such networks are often called multilayer perceptrons (MLPs).
Why multilayer perceptrons?

According to the Universal approximation theorem, a feed-forward neural network
with just a single hidden layer containing a finite number of neurons can approx-
imate a continuous multidimensional function to arbitrary accuracy, assuming
the activation function for the hidden layer is a non-constant, bounded and
monotonically-increasing continuous function.
Note that the requirements on the activation function only applies to the
hidden layer, the output nodes are always assumed to be linear, so as to not
restrict the range of output values.
Mathematical model
The output y is produced via the activation function f
n
!
X
y=f wi xi + bi = f (z),
i=1
Pn
This function receives xi as inputs. Here the activation z = ( i=1 wi xi + bi ).
In an FFNN of such neurons, the inputs xi are the outputs of the neurons in the
preceding layer. Furthermore, an MLP is fully-connected, which means that each
neuron receives a weighted sum of the outputs of all neurons in the previous
layer.
Mathematical model
First, for each node i in the first hidden layer, we calculate a weighted sum zi1 of
the input coordinates xj ,
M
X
zi1 = 1
wij xj + b1i (2)
j=1
Here bi is the so-called bias which is normally needed in case of zero activation
weights or inputs. How to fix the biases and the weights will be discussed below.
4
The value of zi1 is the argument to the activation function fi of each node i, The
variable M stands for all possible inputs to a given node i in the first layer. We
define the output yi1 of all neurons in layer 1 as
 
XM
1 1 1 1
yi = f (zi ) = f  wij xj + bi  (3)
j=1
where we assume that all nodes in the same layer have identical activation
functions, hence the notation f . In general, we could assume in the more general
case that different layers have different activation functions. In this case we
would identify these functions with a superscript l for the l-th layer,
 
Nl−1
X
l l−1
yil = f l (uli ) = f l  wij yj + bli  (4)
j=1
where Nl is the number of nodes in layer l. When the output of all the nodes
in the first hidden layer are computed, the values of the subsequent layer can be
calculated and so forth until the output is obtained.
Mathematical model
The output of neuron i in layer 2 is thus,
 
XN
yi2 = f 2  2 1
wij yj + b2i  (5)
j=1
 ! 
XN M
X
= f2  2 1
wij f 1
wjk xk + b1j + b2i  (6)
j=1 k=1
where we have substituted yk1 with the inputs xk . Finally, the ANN output reads
 
XN
yi3 = f 3  3 2
wij yj + b3i  (7)
j=1
 ! ! 
X X X
3 2 2 1 1
= f3  wij f wjk f wkm xm + b1k + b2j + b31  (8)
j k m
Mathematical model
We can generalize this expression to an MLP with l hidden layers. The complete
functional form is,
5
   
Nl Nl−1 N0
! !
X X X
yil+1 =f l+1  3 l
wij f wl−1
jk . . . f1 1
wmn xn + b1m ... + b2k  + b31 
j=1 k=1 n=1
(9)
which illustrates a basic property of MLPs: The only independent variables

are the input values xn .
Mathematical model
This confirms that an MLP, despite its quite convoluted mathematical form, is
nothing more than an analytic function, specifically a mapping of real-valued
vectors x̂ ∈ Rn → ŷ ∈ Rm .
Furthermore, the flexibility and universality of an MLP can be illustrated
by realizing that the expression is essentially a nested sum of scaled activation
functions of the form
f (x) = c1 f (c2 x + c3 ) + c4 (10)

where the parameters ci are weights and biases. By adjusting these param-
eters, the activation functions can be shifted up and down or left and right,
change slope or be rescaled which is the key to the flexibility of a neural network.
Matrix-vector notation. We can introduce a more convenient notation for

the activations in an A NN.
Additionally, we can represent the biases and activations as layer-wise column
vectors b̂l and ŷl , so that the i-th element of each vector is the bias bli and
activation yil of node i in layer l respectively.
We have that Wl is an Nl−1 × Nl matrix, while b̂l and ŷl are Nl × 1 column
vectors. With this notation, the sum becomes a matrix-vector multiplication,
and we can write the equation for the activations of hidden layer 2 (assuming
three nodes for simplicity) as
 2 2 2
  1   2 
w11 w12 w13 y1 b1
2 2 2   1 
ŷ2 = f2 (W2 ŷ1 + b̂2 ) = f2  w21 w22 w23 · y2 +  b22  . (11)
2 2 2
w31 w32 w33 y31 b23
Matrix-vector notation and activation. The activation of node i in layer

2 is
 
3
X
yi2 = f2 wi1
2 1 2 1
y1 + wi2 2 1
y2 + wi3 y3 + b2i = f2  2 1
wij yj + b2i  . (12)
j=1
This is not just a convenient and compact notation, but also a useful and
intuitive way to think about MLPs: The output is calculated by a series of
6
matrix-vector multiplications and vector additions that are used as input to the
activation functions. For each operation Wl ŷl−1 we move forward one layer.
Activation functions. A property that characterizes a neural network, other

than its connectivity, is the choice of activation function(s). As described in, the
following restrictions are imposed on an activation function for a FFNN to fulfill
the universal approximation theorem
• Non-constant
• Bounded
• Monotonically-increasing
• Continuous
Activation functions, Logistic and Hyperbolic ones. The second require-

ment excludes all linear functions. Furthermore, in a MLP with only linear
activation functions, each layer simply performs a linear transformation of its
inputs.
Regardless of the number of layers, the output of the NN will be nothing
but a linear function of the inputs. Thus we need to introduce some kind of
non-linearity to the NN to be able to fit non-linear functions Typical examples
are the logistic Sigmoid
1
f (x) = ,
1 + e−x
and the hyperbolic tangent function
f (x) = tanh(x)
Relevance. The sigmoid function are more biologically plausible because the
output of inactive neurons are zero. Such activation function are called one-sided.
However, it has been shown that the hyperbolic tangent performs better than
the sigmoid for training MLPs. has become the most popular for deep neural
networks
"""The sigmoid function (or the logistic curve) is a
function that takes any real number, z, and outputs a number (0,1).
It is useful in neural networks for assigning weights on a relative scale.
The value z is the weighted sum of parameters involved in the learning algorithm."""
import numpy
import matplotlib.pyplot as plt
import math as mt
z = numpy.arange(-5, 5, .1)
sigma_fn = numpy.vectorize(lambda z: 1/(1+numpy.exp(-z)))
sigma = sigma_fn(z)
7
fig = plt.figure()
ax = fig.add_subplot(111)
ax.plot(z, sigma)
ax.set_ylim([-0.1, 1.1])
ax.set_xlim([-5,5])
ax.grid(True)
ax.set_xlabel(’z’)
ax.set_title(’sigmoid function’)
plt.show()
"""Step Function"""
step_fn = numpy.vectorize(lambda z: 1.0 if z >= 0.0 else 0.0)
step = step_fn(z)
fig = plt.figure()
ax.plot(z, step)
ax.set_ylim([-0.5, 1.5])
ax.set_xlim([-5,5])
ax.grid(True)
ax.set_title(’step function’)
plt.show()
"""Sine Function"""
z = numpy.arange(-2*mt.pi, 2*mt.pi, 0.1)
t = numpy.sin(z)
fig = plt.figure()
ax.plot(z, t)
ax.set_ylim([-1.0, 1.0])
ax.set_xlim([-2*mt.pi,2*mt.pi])
ax.grid(True)
ax.set_title(’sine function’)
plt.show()
"""Plots a graph of the squashing function used by a rectified linear

unit"""
zero = numpy.zeros(len(z))
y = numpy.max([zero, z], axis=0)
fig = plt.figure()
ax.plot(z, y)
ax.set_ylim([-2.0, 2.0])
ax.set_xlim([-2.0, 2.0])
ax.grid(True)
ax.set_title(’Rectified linear unit’)
plt.show()
8
The multilayer perceptron (MLP)
The multilayer perceptron is a very popular, and easy to implement approach,
to deep learning. It consists of
1. A neural network with one or more layers of nodes between the input and
the output nodes.
2. The multilayer network structure, or architecture, or topology, consists of
an input layer, one or more hidden layers, and one output layer.
3. The input nodes pass values to the first hidden layer, its nodes pass the
information on to the second and so on till we reach the output layer.
As a convention it is normal to call a network with one layer of input units, one
layer of hidden units and one layer of output units as a two-layer network. A
network with two layers of hidden units is called a three-layer network etc etc.
For an MLP network there is no direct connection between the output
nodes/neurons/units and the input nodes/neurons/units. Hereafter we will call
the various entities of a layer for nodes. There are also no connections within a
single layer.
The number of input nodes does not need to equal the number of output
nodes. This applies also to the hidden layers. Each layer may have its own
number of nodes and activation functions.
The hidden layers have their name from the fact that they are not linked to
observables and as we will see below when we define the so-called activation ẑ,
we can think of this as a basis expansion of the original inputs x̂. The difference
however between neural networks and say linear regression is that now these
basis functions (which will correspond to the weights in the network) are learned
from data. This results in an important difference between neural networks and
deep learning approaches on one side and methods like logistic regression or
linear regression and their modifications on the other side.
From one to many layers, the universal approximation the-

orem
A neural network with only one layer, what we called the simple perceptron, is
best suited if we have a standard binary model with clear (linear) boundaries
between the outcomes. As such it could equally well be replaced by standard
linear regression or logistic regression. Networks with one or more hidden layers
approximate systems with more complex boundaries.
As stated earlier, an important theorem in studies of neural networks, restated
without proof here, is the universal approximation theorem.
It states that a feed-forward network with a single hidden layer containing
a finite number of neurons can approximate continuous functions on compact
subsets of real functions. The theorem thus states that simple neural networks
can represent a wide variety of interesting functions when given appropriate
9
parameters. It is the multilayer feedforward architecture itself which gives neural
networks the potential of being universal approximators.
Deriving the back propagation code for a multilayer per-

ceptron model
As we have seen now in a feed forward network, we can express the final output
of our network in terms of basic matrix-vector multiplications. The unknowwn
quantities are our weights wij and we need to find an algorithm for changing
them so that our errors are as small as possible. This leads us to the famous
back propagation algorithm.
The questions we want to ask are how do changes in the biases and the
weights in our network change the cost function and how can we use the final
output to modify the weights?
To derive these equations let us start with a plain regression problem and
define our cost function as
n
1X 2
C(Ŵ ) = (yi − ti ) ,
2 i=1
where the ti s are our n targets (the values we want to reproduce), while
the outputs of the network after having propagated all inputs x̂ are given by
yi . Below we will demonstrate how the basic equations arising from the back
propagation algorithm can be modified in order to study classification problems
with K classes.
Definitions
With our definition of the targets t̂, the outputs of the network ŷ and the inputs
x̂ we define now the activation zjl of node/neuron/unit j of the l-th layer as a
function of the bias, the weights which add up from the previous layer l − 1 and
the forward passes/outputs âl−1 from the previous layer as
Ml−1
X
l l−1
zjl = wij ai + blj ,
i=1
where blkare the biases from layer l. Here Ml−1 represents the total number
of nodes/neurons/units of layer l − 1. The figure here illustrates this equation.
We can rewrite this in a more compact form as the matrix-vector products we
discussed earlier,
T
ẑ l = Ŵ l âl−1 + b̂l .
With the activation values ẑ l we can in turn define the output of layer l as
â = f (ẑ l ) where f is our activation function. In the examples here we will use
l
the sigmoid function discussed in our logistic regression lectures. We will also
10
use the same activation function f for all layers and their nodes. It means we
have
1
alj = f (zjl ) = .
1 + exp −(zjl )
Derivatives and the chain rule

From the definition of the activation zjl we have
∂zjl
l
= al−1
i ,
∂wij
and
∂zjl l
= wji .
∂al−1
i
With our definition of the activation function we have that (note that this
function depends only on zjl )
∂alj
= alj (1 − alj ) = f (zjl )(1 − f (zjl )).
∂zjl
Derivative of the cost function

With these definitions we can now compute the derivative of the cost function in
terms of the weights.
Let us specialize to the output layer l = L. Our cost function is
n n
1X 1X L 2
C(WˆL ) =
2
(yi − ti ) = ai − ti ,
2 i=1 2 i=1
The derivative of this function with respect to the weights is
∂C(WˆL ) L
∂aL
j
L
= aj − tj L
,
∂wjk ∂wjk
The last partial derivative can easily be computed and reads (by applying the
chain rule)
∂aL
j ∂aLj ∂zj
L
L L−1
L
= = aLj (1 − aj )ak ,
∂wjk ∂zjL ∂wjk
L
Bringing it together, first back propagation equation

We have thus
∂C(WˆL ) L L−1
= aL
L
L j − tj aj (1 − aj )ak ,
∂wjk
11
Defining
0 L ∂C
δjL = aL L L

j (1 − aj ) aj − tj = f (zj ) ,
∂(aL
j)
and using the Hadamard product of two vectors we can write this as
∂C
δ̂ L = f 0 (ẑ L ) ◦ .
∂(âL )
This is an important expression. The second term on the right handside

measures how fast the cost function is changing as a function of the jth output
activation. If, for example, the cost function doesn’t depend much on a particular
output node j, then δjL will be small, which is what we would expect. The first
term on the right, measures how fast the activation function f is changing at a
given activation value zjL .
Notice that everything in the above equations is easily computed. In particu-
lar, we compute zjL while computing the behaviour of the network, and it is only
a small additional overhead to compute f 0 (zjL ). The exact form of the derivative
with respect to the output depends on the form of the cost function. However,
provided the cost function is known there should be little trouble in calculating
∂C
∂(aL
j)
With the definition of δjL we have a more compact definition of the derivative
of the cost function in terms of the weights, namely
∂C(WˆL )
L
= δjL akL−1 .
∂wjk
Derivatives in terms of zjL

It is also easy to see that our previous equation can be written as
∂C ∂C ∂aLj
δjL = = ,
∂zjL ∂aL L
j ∂zj
which can also be interpreted as the partial derivative of the cost function with
respect to the biases bL
j , namely
∂C ∂bLj ∂C
δjL = L L
= L,
∂bj ∂zj ∂bj
That is, the error δjL is exactly equal to the rate of change of the cost function
as a function of the bias.
12
Bringing it together
We have now three equations that are essential for the computations of the
derivatives of the cost function at the output layer. These equations are needed
to start the algorithm and they are
The starting equations.
∂C(WˆL )
L
= δjL akL−1 , (13)
∂wjk
and
∂C
δjL = f 0 (zjL ) , (14)
∂(aL
j)
and
∂C
δjL = , (15)
∂bL
j
An interesting consequence of the above equations is that when the activation

aL−1
k is small, the gradient term, that is the derivative of the cost function with
respect to the weights, will also tend to be small. We say then that the weight
learns slowly, meaning that it changes slowly when we minimize the weights via
say gradient descent. In this case we say the system learns slowly.
Another interesting feature is that is when the activation function, represented
by the sigmoid function here, is rather flat when we move towards its end values
0 and 1 (see the above Python codes). In these cases, the derivatives of the
activation function will also be close to zero, meaning again that the gradients
will be small and the network learns slowly again.
We need a fourth equation and we are set. We are going to propagate
backwards in order to the determine the weights and biases. In order to do so
we need to represent the error in the layer before the final one L − 1 in terms of
the errors in the final output layer.
Final back propagating equation

We have that (replacing L with a general layer l)
∂C
δjl = .
∂zjl
We want to express this in terms of the equations for layer l + 1. Using the chain
rule and summing over all k entries we have
X ∂C ∂z l+1 l+1
l+1 ∂zk
X
δjl = k
= δk ,
k
∂zkl+1 ∂zjl k
∂zjl
13
and recalling that
Ml
X
zjl+1 = l+1 l
wij ai + bl+1
j ,
i=1
with Ml being the number of nodes in layer l, we obtain

X
l+1 0 l
δjl = δkl+1 wkj f (zj ),
k
This is our final equation.

We are now ready to set up the algorithm for back propagation and learning
the weights and biases.
Setting up the Back propagation algorithm

The four equations provide us with a way of computing the gradient of the cost
function. Let us write this out in the form of an algorithm.
First, we set up the input data x̂ and the activations ẑ1 of the input layer and
compute the activation function and the pertinent outputs â1 .
Secondly, we perform then the feed forward till we reach the output layer and
compute all ẑl of the input layer and compute the activation function and the
pertinent outputs âl for l = 2, 3, . . . , L.
Thereafter we compute the ouput error δ̂ L by computing all

∂C
δjL = f 0 (zjL ) .
∂(aL
j)
Then we compute the back propagate error for each l = L − 1, L − 2, . . . , 2 as

X
l+1 0 l
δjl = δkl+1 wkj f (zj ).
k
Finally, we update the weights and the biases using gradient descent for each
l = L − 1, L − 2, . . . , 2 and update the weights and biases according to the rules
l
wjk l
←= wjk − ηδjl al−1
k ,
∂C
blj ← blj − η = blj − ηδjl ,
∂blj
The parameter η is the learning parameter discussed in connection with
the gradient descent methods. Here it is convenient to use stochastic gradient
descent (see the examples below) with mini-batches with an outer loop that
steps through multiple epochs of training.
14
Setting up a Multi-layer perceptron model for classification
We are now gong to develop an example based on the MNIST data base. This
is a classification problem and we need to use our cross-entropy function we
discussed in connection with logistic regression. The cross-entropy defines our
cost function for the classificaton problems with neural networks.
In binary classification with two classes (0, 1) we define the logistic/sigmoid
function as the probability that a particular input is in class 0 or 1. This is
possible because the logistic function takes any input from the real numbers
and inputs a number between 0 and 1, and can therefore be interpreted as a
probability. It also has other nice properties, such as a derivative that is simple
to calculate.
For an input a from the hidden layer, the probability that the input x is in
class 0 or 1 is just. We let θ represent the unknown weights and biases to be
adjusted by our equations). The variable x represents our activation values z.
We have
1
P (y = 0 | x̂, θ̂) = ,
1 + exp (−x̂)
and
P (y = 1 | x̂, θ̂) = 1 − P (y = 0 | x̂, θ̂),
where y ∈ {0, 1} and θ̂ represents the weights and biases of our network.
Defining the cost function

Our cost function is given as (see the Logistic regression lectures)
n
X n
X
C(θ̂) = − ln P (D | θ̂) = − yi ln[P (yi = 0)]+(1−yi ) ln[1−P (yi = 0)] = Li (θ̂).
i=1 i=1
This last equality means that we can interpret our cost function as a sum
over the loss function for each point in the dataset Li (θ̂). The negative sign is
just so that we can think about our algorithm as minimizing a positive number,
rather than maximizing a negative number.
In multiclass classification it is common to treat each integer label as a so
called one-hot vector:
y = 5 → ŷ = (0, 0, 0, 0, 0, 1, 0, 0, 0, 0), and
y = 1 → ŷ = (0, 1, 0, 0, 0, 0, 0, 0, 0, 0),
i.e. a binary bit string of length C, where C = 10 is the number of classes in
the MNIST dataset (numbers from 0 to 9)..
If x̂i is the i-th input (image), yic refers to the c-th component of the i-th
output vector ŷi . The probability of x̂i being in class c will be given by the
softmax function:
exp ((âhidden
i )T ŵc )
P (yic = 1 | x̂i , θ̂) = PC−1 ,
hidden )T ŵ 0 )
c0 =0 exp ((âi c
15
which reduces to the logistic function in the binary case. The likelihood of
this C-class classifier is now given as:
n C−1
Y Y
P (D | θ̂) = [P (yic = 1)]yic .
i=1 c=0
Again we take the negative log-likelihood to define our cost function:
C(θ̂) = − log P (D | θ̂).

See the logistic regression lectures for a full definition of the cost function.
The back propagation equations need now only a small change, namely the
definition of a new cost function. We are thus ready to use the same equations
as before!
Example: binary classification problem

As an example of the above, relevant for project 2 as well, let us consider a
binary class. As discussed in our logistic regression lectures, we defined a cost
function in terms of the parameters β as
n
X
C(β̂) = − yi log p(yi |xi , β̂) + (1 − yi ) log 1 − p(yi |xi , β̂) ,
i=1
where we had defined the logistic (sigmoid) function
exp (β0 + β1 xi )
p(yi = 1|xi , β̂) = ,
1 + exp (β0 + β1 xi )
and
p(yi = 0|xi , β̂) = 1 − p(yi = 1|xi , β̂).
The parameters β̂ were defined using a minimization method like gradient descent
or Newton-Raphson’s method.
Now we replace xi with the activation zil for a given layer l and the outputs
as yi = ali = f (zil ), with zil now being a function of the weights wij
l
and biases bli .
We have then
exp (zil )
ali = yi = ,
1 + exp (zil )
with X
l l−1
zil = wij aj + bli ,
j
where the superscript l − 1 indicates that these are the outputs from layer l − 1.
Our cost function at the final layer l = L is now
n
X
ti log aL L

C(Ŵ ) = − i + (1 − ti ) log (1 − ai ) ,
i=1
16
where we have defined the targets ti . The derivatives of the cost function with
respect to the output aL
i are then easily calculated and we get
∂C(Ŵ ) aL
i − ti
= .
∂aL
i aL (1 − aL )
i i
In case we use another activation function than the logistic one, we need to
evaluate other derivatives.
The Softmax function

In case we employ the more general case given by the Softmax equation, we
need to evaluate the derivative of the activation function with respect to the
activation zil , that is we need
∂f (zil ) ∂f (zil ) ∂zjl ∂f (zil ) l−1

= = a .
l
∂wjk l
∂zj ∂wjk l ∂zjl k
For the Softmax function we have

exp (zil )
f (zil ) = PK .
l
m=1 exp (zm )
Its derivative with respect to zjl gives
∂f (zil )
= f (zil ) δij − f (zjl ) ,

∂zjl
which in case of the simply binary model reduces to having i = j.
Developing a code for doing neural networks with back

propagation
One can identify a set of key steps when using neural networks to solve supervised
learning problems:
1. Collect and pre-process data
2. Define model and architecture

3. Choose cost function and optimizer
4. Train the model
5. Evaluate model performance on test data

6. Adjust hyperparameters (if necessary, network architecture)
17
Collect and pre-process data
Here we will be using the MNIST dataset, which is readily available through
the scikit-learn package. You may also find it for example here. The MNIST
(Modified National Institute of Standards and Technology) database is a large
database of handwritten digits that is commonly used for training various image
processing systems. The MNIST dataset consists of 70 000 images of size 28 × 28
pixels, each labeled from 0 to 9. The scikit-learn dataset we will use consists of a
selection of 1797 images of size 8 × 8 collected and processed from this database.
To feed data into a feed-forward neural network we need to represent the
inputs as a design/feature matrix X = (ninputs , nf eatures ). Each row represents
an input, in this case a handwritten digit, and each column represents a feature,
in this case a pixel. The correct answers, also known as labels or targets are
represented as a 1D array of integers Y = (ninputs ) = (5, 3, 1, 8, ...).
As an example, say we want to build a neural network using supervised
learning to predict Body-Mass Index (BMI) from measurements of height (in
m) and weight (in kg). If we have measurements of 5 people the design/feature
matrix could be for example:
 
1.85&81
 1.71&65 
 
X= 1.95&103 ,

 1.55&42 
1.63&56
and the targets would be:
Y = (23.7, 22.2, 27.1, 17.5, 21.1)

Since each input image is a 2D matrix, we need to flatten the image (i.e.
"unravel" the 2D matrix into a 1D array) to turn the data into a design/feature
matrix. This means we lose all spatial information in the image, such as
locality and translational invariance. More complicated architectures such as
Convolutional Neural Networks can take advantage of such information, and are
most commonly applied when analyzing images.
# import necessary packages
import numpy as np
from sklearn import datasets
# ensure the same random numbers appear every time

np.random.seed(0)
# display images in notebook

%matplotlib inline
plt.rcParams[’figure.figsize’] = (12,12)
# download MNIST dataset

digits = datasets.load_digits()
18
# define inputs and labels
inputs = digits.images
labels = digits.target
print("inputs = (n_inputs, pixel_width, pixel_height) = " + str(inputs.shape))
print("labels = (n_inputs) = " + str(labels.shape))
# flatten the image

# the value -1 means dimension is inferred from the remaining dimensions: 8x8 = 64
n_inputs = len(inputs)
inputs = inputs.reshape(n_inputs, -1)
print("X = (n_inputs, n_features) = " + str(inputs.shape))
# choose some random images to display

indices = np.arange(n_inputs)
random_indices = np.random.choice(indices, size=5)
for i, image in enumerate(digits.images[random_indices]):

plt.subplot(1, 5, i+1)
plt.axis(’off’)
plt.imshow(image, cmap=plt.cm.gray_r, interpolation=’nearest’)
plt.title("Label: %d" % digits.target[random_indices[i]])
plt.show()
Train and test datasets

Performing analysis before partitioning the dataset is a major error, that can
lead to incorrect conclusions.
We will reserve 80% of our dataset for training and 20% for testing.
It is important that the train and test datasets are drawn randomly from our
dataset, to ensure no bias in the sampling. Say you are taking measurements
of weather data to predict the weather in the coming 5 days. You don’t want
to train your model on measurements taken from the hours 00.00 to 12.00, and
then test it on data collected from 12.00 to 24.00.
from sklearn.model_selection import train_test_split
# one-liner from scikit-learn library

train_size = 0.8
test_size = 1 - train_size
X_train, X_test, Y_train, Y_test = train_test_split(inputs, labels, train_size=train_size,
test_size=test_size)
# equivalently in numpy
def train_test_split_numpy(inputs, labels, train_size, test_size):
inputs_shuffled = inputs.copy()
labels_shuffled = labels.copy()
np.random.shuffle(inputs_shuffled)
np.random.shuffle(labels_shuffled)
train_end = int(n_inputs*train_size)
X_train, X_test = inputs_shuffled[:train_end], inputs_shuffled[train_end:]
19
Y_train, Y_test = labels_shuffled[:train_end], labels_shuffled[train_end:]
return X_train, X_test, Y_train, Y_test
#X_train, X_test, Y_train, Y_test = train_test_split_numpy(inputs, labels, train_size, test_size)
print("Number of training images: " + str(len(X_train)))

print("Number of test images: " + str(len(X_test)))
Define model and architecture

Our simple feed-forward neural network will consist of an input layer, a single
hidden layer and an output layer. The activation y of each neuron is a weighted
sum of inputs, passed through an activation function. In case of the simple
perceptron model we have
n
X
z= wi ai ,
i=1
y = f (z),
where f is the activation function, ai represents input from neuron i in the
preceding layer and wi is the weight to input i. The activation of the neurons in
the input layer is just the features (e.g. a pixel value).
The simplest activation function for a neuron is the Heaviside function:
(
1, z > 0
f (z) =
0, otherwise
A feed-forward neural network with this activation is known as a perceptron.
For a binary classifier (i.e. two classes, 0 or 1, dog or not-dog) we can also use this
in our output layer. This activation can be generalized to k classes (using e.g. the
one-against-all strategy), and we call these architectures multiclass perceptrons.
However, it is now common to use the terms Single Layer Perceptron (SLP)
(1 hidden layer) and Multilayer Perceptron (MLP) (2 or more hidden layers) to
refer to feed-forward neural networks with any activation function.
Typical choices for activation functions include the sigmoid function, hyper-
bolic tangent, and Rectified Linear Unit (ReLU). We will be using the sigmoid
function σ(x):
1
f (x) = σ(x) = ,
1 + e−x
which is inspired by probability theory (see logistic regression) and was most
commonly used until about 2011. See the discussion below concerning other
activation functions.
20
Layers
• Input
Since each input image has 8x8 = 64 pixels or features, we have an input layer
of 64 neurons.
• Hidden layer
We will use 50 neurons in the hidden layer receiving input from the neurons in
the input layer. Since each neuron in the hidden layer is connected to the 64
inputs we have 64x50 = 3200 weights to the hidden layer.
• Output
If we were building a binary classifier, it would be sufficient with a single neuron

in the output layer, which could output 0 or 1 according to the Heaviside function.
This would be an example of a hard classifier, meaning it outputs the class of the
input directly. However, if we are dealing with noisy data it is often beneficial
to use a soft classifier, which outputs the probability of being in class 0 or 1.
For a soft binary classifier, we could use a single neuron and interpret the
output as either being the probability of being in class 0 or the probability of
being in class 1. Alternatively we could use 2 neurons, and interpret each neuron
as the probability of being in each class.
Since we are doing multiclass classification, with 10 categories, it is natural to
use 10 neurons in the output layer. We number the neurons j = 0, 1, ..., 9. The
activation of each output neuron j will be according to the softmax function:
exp (âT ŵj )

P (class j | input â) = P9 ,
T
c=0 exp (â ŵc )
i.e. each neuron j outputs the probability of being in class j given an input
from the hidden layer â, with ŵj the weights of neuron j to the inputs. The
denominator is a normalization factor to ensure the outputs (probabilities) sum
up to 1. The exponent is just the weighted sum of inputs as before:
n
X
zj = wij ai + bj .
i=1
Since each neuron in the output layer is connected to the 50 inputs from the
hidden layer we have 50x10 = 500 weights to the output layer.
Weights and biases

Typically weights are initialized with small values distributed around zero, drawn
from a uniform or normal distribution. Setting all weights to zero means all
neurons give the same output, making the network useless.
Adding a bias value to the weighted sum of inputs allows the neural network
to represent a greater range of values. Without it, any input with the value 0
21
will be mapped to zero (before being passed through the activation). The bias
unit has an output of 1, and a weight to each neuron j, bj :
n
X
zj = wij ai + bj .
i=1
The bias weights b̂ are often initialized to zero, but a small value like 0.01
ensures all neurons have some output which can be backpropagated in the first
training cycle.
# building our neural network
n_inputs, n_features = X_train.shape

n_hidden_neurons = 50
n_categories = 10
# we make the weights normally distributed using numpy.random.randn
# weights and bias in the hidden layer

hidden_weights = np.random.randn(n_features, n_hidden_neurons)
hidden_bias = np.zeros(n_hidden_neurons) + 0.01
# weights and bias in the output layer

output_weights = np.random.randn(n_hidden_neurons, n_categories)
output_bias = np.zeros(n_categories) + 0.01
Feed-forward pass
Denote F the number of features, H the number of hidden neurons and C the
number of categories. For each input image we calculate a weighted sum of input
features (pixel values) to each neuron j in the hidden layer l:
F
X
zjl = l
wij xi + blj ,
i=1
this is then passed through our activation function
alj = f (zjl ).
We calculate a weighted sum of inputs (activations in the hidden layer) to
each neuron j in the output layer:
H
X
zjL = L l
wij ai + bL
j.
i=1
Finally we calculate the output of neuron j in the output layer using the
softmax function:
exp (zjL )
aL
j = PC−1 .
L
c=0 exp (zc )
22
Matrix multiplications
Since our data has the dimensions X = (ninputs , nf eatures ) and our weights to
the hidden layer have the dimensions Whidden = (nf eatures , nhidden ), we can
easily feed the network all our training data in one go by taking the matrix
product
XW h = (ninputs , nhidden ),
and obtain a matrix that holds the weighted sum of inputs to the hidden
layer for each input image and each hidden neuron. We also add the bias to
obtain a matrix of weighted sums to the hidden layer Z h :
ẑ l = X̂ Ŵ l + b̂l ,
meaning the same bias (1D array with size equal number of hidden neurons)
is added to each input image. This is then passed through the activation:
âl = f (ẑ l ).
This is fed to the output layer:
ẑ L = âL Ŵ L + b̂L .
Finally we receive our output values for each image and each category by
passing it through the softmax function:
output = sof tmax(ẑ L ) = (ninputs , ncategories ).

# setup the feed-forward pass, subscript h = hidden layer
def sigmoid(x):
return 1/(1 + np.exp(-x))
def feed_forward(X):
# weighted sum of inputs to the hidden layer
z_h = np.matmul(X, hidden_weights) + hidden_bias
# activation in the hidden layer
a_h = sigmoid(z_h)
# weighted sum of inputs to the output layer

z_o = np.matmul(a_h, output_weights) + output_bias
# softmax output
# axis 0 holds each input and axis 1 the probabilities of each category
exp_term = np.exp(z_o)
probabilities = exp_term / np.sum(exp_term, axis=1, keepdims=True)
return probabilities
probabilities = feed_forward(X_train)
print("probabilities = (n_inputs, n_categories) = " + str(probabilities.shape))
print("probability that image 0 is in category 0,1,2,...,9 = \n" + str(probabilities[0]))
print("probabilities sum up to: " + str(probabilities[0].sum()))
print()
23
# we obtain a prediction by taking the class with the highest likelihood
def predict(X):
probabilities = feed_forward(X)
return np.argmax(probabilities, axis=1)
predictions = predict(X_train)
print("predictions = (n_inputs) = " + str(predictions.shape))
print("prediction for image 0: " + str(predictions[0]))
print("correct label for image 0: " + str(Y_train[0]))
Choose cost function and optimizer

To measure how well our neural network is doing we need to introduce a cost
function. We will call the function that gives the error of a single sample output
the loss function, and the function that gives the total error of our network
across all samples the cost function. A typical choice for multiclass classification
is the cross-entropy loss, also known as the negative log likelihood.
In multiclass classification it is common to treat each integer label as a so
called one-hot vector:
y=5 → ŷ = (0, 0, 0, 0, 0, 1, 0, 0, 0, 0),
y=1 → ŷ = (0, 1, 0, 0, 0, 0, 0, 0, 0, 0),

i.e. a binary bit string of length C, where C = 10 is the number of classes in
the MNIST dataset.
Let yic denote the c-th component of the i-th one-hot vector. We define the
cost function C as a sum over the cross-entropy loss for each point x̂i in the
dataset.
In the one-hot representation only one of the terms in the loss function is
non-zero, namely the probability of the correct category c0 (i.e. the category c0
such that yic0 = 1). This means that the cross entropy loss only punishes you
for how wrong you got the correct label. The probability of category c is given
by the softmax function. The vector θ̂ represents the parameters of our network,
i.e. all the weights and biases.
Optimizing the cost function

The network is trained by finding the weights and biases that minimize the cost
function. One of the most widely used classes of methods is gradient descent
and its generalizations. The idea behind gradient descent is simply to adjust
the weights in the direction where the gradient of the cost function is large and
negative. This ensures we flow toward a local minimum of the cost function.
Each parameter θ is iteratively adjusted according to the rule
θi+1 = θi − η∇C(θi ),
24
where η is known as the learning rate, which controls how big a step we take
towards the minimum. This update can be repeated for any number of iterations,
or until we are satisfied with the result.
A simple and effective improvement is a variant called Batch Gradient De-
scent. Instead of calculating the gradient on the whole dataset, we calculate an
approximation of the gradient on a subset of the data called a minibatch. If
there are N data points and we have a minibatch size of M , the total number
of batches is N/M . We denote each minibatch Bk , with k = 1, 2, ..., N/M . The
gradient then becomes:
N
1 X 1 X
∇C(θ) = ∇Li (θ) → ∇Li (θ),
N i=1 M
i∈Bk
i.e. instead of averaging the loss over the entire dataset, we average over a
minibatch.
This has two important benefits:
1. Introducing stochasticity decreases the chance that the algorithm becomes
stuck in a local minima.
2. It significantly speeds up the calculation, since we do not have to use the
entire dataset to calculate the gradient.
The various optmization methods, with codes and algorithms, are discussed in
our lectures on Gradient descent approaches.
Regularization
It is common to add an extra term to the cost function, proportional to the size
of the weights. This is equivalent to constraining the size of the weights, so that
they do not grow out of control. Constraining the size of the weights means that
the weights cannot grow arbitrarily large to fit the training data, and in this
way reduces overfitting.
We will measure the size of the weights using the so called L2-norm, meaning
our cost function becomes:
N N N
1 X 1 X 1 X X
C(θ) = Li (θ) → Li (θ) + λ||ŵ||22 = L(θ) + λ 2
wij ,
N i=1 N i=1 N i=1 ij
i.e. we sum up all the weights squared. The factor λ is known as a regular-
ization parameter.
In order to train the model, we need to calculate the derivative of the cost
function with respect to every bias and weight in the network. In total our
network has (64+1)×50 = 3250 weights in the hidden layer and (50+1)×10 = 510
weights to the output layer (+1 for the bias), and the gradient must be calculated
for every parameter. We use the backpropagation algorithm discussed above.
This is a clever use of the chain rule that allows us to calculate the gradient
efficently.
25
Matrix multiplication
To more efficently train our network these equations are implemented using
matrix operations. The error in the output layer is calculated simply as, with t̂
being our targets,
δL = t̂ − ŷ = (ninputs , ncategories ).
The gradient for the output weights is calculated as
∇WL = âT δL = (nhidden , ncategories ),

where â = (ninputs , nhidden ). This simply means that we are summing up
the gradients for each input. Since we are going backwards we have to transpose
the activation matrix.
The gradient with respect to the output bias is then
ninputs
X
∇b̂L = δL = (ncategories ).
i=1
The error in the hidden layer is
∆h = δL WLT ◦ f 0 (zh ) = δL WLT ◦ ah ◦ (1 − ah ) = (ninputs , nhidden ),
where f 0 (ah ) is the derivative of the activation in the hidden layer. The
matrix products mean that we are summing up the products for each neuron
in the output layer. The symbol ◦ denotes the Hadamard product, meaning
element-wise multiplication.
This again gives us the gradients in the hidden layer:
∇Wh = X T δh = (nf eatures , nhidden ),
ninputs
X
∇bh = δh = (nhidden ).
i=1
# to categorical turns our integer vector into a onehot representation

from sklearn.metrics import accuracy_score
# one-hot in numpy
def to_categorical_numpy(integer_vector):
n_inputs = len(integer_vector)
n_categories = np.max(integer_vector) + 1
onehot_vector = np.zeros((n_inputs, n_categories))
onehot_vector[range(n_inputs), integer_vector] = 1
return onehot_vector
#Y_train_onehot, Y_test_onehot = to_categorical(Y_train), to_categorical(Y_test)
Y_train_onehot, Y_test_onehot = to_categorical_numpy(Y_train), to_categorical_numpy(Y_test)
26
def feed_forward_train(X):
# weighted sum of inputs to the hidden layer
z_h = np.matmul(X, hidden_weights) + hidden_bias
# activation in the hidden layer
a_h = sigmoid(z_h)
# weighted sum of inputs to the output layer

z_o = np.matmul(a_h, output_weights) + output_bias
# softmax output
# axis 0 holds each input and axis 1 the probabilities of each category
# for backpropagation need activations in hidden and output layers
return a_h, probabilities
def backpropagation(X, Y):

a_h, probabilities = feed_forward_train(X)
# error in the output layer

error_output = probabilities - Y
# error in the hidden layer
error_hidden = np.matmul(error_output, output_weights.T) * a_h * (1 - a_h)
# gradients for the output layer

output_weights_gradient = np.matmul(a_h.T, error_output)
output_bias_gradient = np.sum(error_output, axis=0)
# gradient for the hidden layer

hidden_weights_gradient = np.matmul(X.T, error_hidden)
hidden_bias_gradient = np.sum(error_hidden, axis=0)
return output_weights_gradient, output_bias_gradient, hidden_weights_gradient, hidden_bias_gra

print("Old accuracy on training data: " + str(accuracy_score(predict(X_train), Y_train)))
eta = 0.01
lmbd = 0.01
for i in range(1000):
# calculate gradients
dWo, dBo, dWh, dBh = backpropagation(X_train, Y_train_onehot)
# regularization term gradients
dWo += lmbd * output_weights
dWh += lmbd * hidden_weights
# update weights and biases

output_weights -= eta * dWo
output_bias -= eta * dBo
hidden_weights -= eta * dWh
hidden_bias -= eta * dBh
print("New accuracy on training data: " + str(accuracy_score(predict(X_train), Y_train)))
Improving performance
As we can see the network does not seem to be learning at all. It seems to be
just guessing the label for each image. In order to obtain a network that does
something useful, we will have to do a bit more work.
27
The choice of hyperparameters such as learning rate and regularization
parameter is hugely influential for the performance of the network. Typically a
grid-search is performed, wherein we test different hyperparameters separated
by orders of magnitude. For example we could test the learning rates η =
10−6 , 10−5 , ..., 10−1 with different regularization parameters λ = 10−6 , ..., 10−0 .
Next, we haven’t implemented minibatching yet, which introduces stochastic-
ity and is though to act as an important regularizer on the weights. We call a
feed-forward + backward pass with a minibatch an iteration, and a full training
period going through the entire dataset (n/M batches) an epoch.
If this does not improve network performance, you may want to consider
altering the network architecture, adding more neurons or hidden layers. Andrew
Ng goes through some of these considerations in this video. You can find a
summary of the video here.
Full object-oriented implementation

It is very natural to think of the network as an object, with specific instances of
the network being realizations of this object with different hyperparameters. An
implementation using Python classes provides a clean structure and interface,
and the full implementation of our neural network is given below.
class NeuralNetwork:
def __init__(
self,
X_data,
Y_data,
n_hidden_neurons=50,
n_categories=10,
epochs=10,
batch_size=100,
eta=0.1,
lmbd=0.0):
self.X_data_full = X_data
self.Y_data_full = Y_data
self.n_inputs = X_data.shape[0]
self.n_features = X_data.shape[1]
self.n_hidden_neurons = n_hidden_neurons
self.n_categories = n_categories
self.epochs = epochs
self.batch_size = batch_size
self.iterations = self.n_inputs // self.batch_size
self.eta = eta
self.lmbd = lmbd
self.create_biases_and_weights()
def create_biases_and_weights(self):
self.hidden_weights = np.random.randn(self.n_features, self.n_hidden_neurons)
self.hidden_bias = np.zeros(self.n_hidden_neurons) + 0.01
self.output_weights = np.random.randn(self.n_hidden_neurons, self.n_categories)

self.output_bias = np.zeros(self.n_categories) + 0.01
28
def feed_forward(self):
# feed-forward for training
self.z_h = np.matmul(self.X_data, self.hidden_weights) + self.hidden_bias
self.a_h = sigmoid(self.z_h)
self.z_o = np.matmul(self.a_h, self.output_weights) + self.output_bias
exp_term = np.exp(self.z_o)
self.probabilities = exp_term / np.sum(exp_term, axis=1, keepdims=True)
def feed_forward_out(self, X):

# feed-forward for output
z_h = np.matmul(X, self.hidden_weights) + self.hidden_bias
a_h = sigmoid(z_h)
z_o = np.matmul(a_h, self.output_weights) + self.output_bias

def backpropagation(self):
error_output = self.probabilities - self.Y_data
error_hidden = np.matmul(error_output, self.output_weights.T) * self.a_h * (1 - self.a_h)
self.output_weights_gradient = np.matmul(self.a_h.T, error_output)
self.output_bias_gradient = np.sum(error_output, axis=0)
self.hidden_weights_gradient = np.matmul(self.X_data.T, error_hidden)
self.hidden_bias_gradient = np.sum(error_hidden, axis=0)
if self.lmbd > 0.0:

self.output_weights_gradient += self.lmbd * self.output_weights
self.hidden_weights_gradient += self.lmbd * self.hidden_weights
self.output_weights -= self.eta * self.output_weights_gradient

self.output_bias -= self.eta * self.output_bias_gradient
self.hidden_weights -= self.eta * self.hidden_weights_gradient
self.hidden_bias -= self.eta * self.hidden_bias_gradient
def predict(self, X):

probabilities = self.feed_forward_out(X)
return np.argmax(probabilities, axis=1)
def predict_probabilities(self, X):
probabilities = self.feed_forward_out(X)
def train(self):
data_indices = np.arange(self.n_inputs)
for i in range(self.epochs):
for j in range(self.iterations):
# pick datapoints with replacement
chosen_datapoints = np.random.choice(
data_indices, size=self.batch_size, replace=False
)
# minibatch training data

self.X_data = self.X_data_full[chosen_datapoints]
29
self.Y_data = self.Y_data_full[chosen_datapoints]
self.feed_forward()
self.backpropagation()
Evaluate model performance on test data

To measure the performance of our network we evaluate how well it does it data
it has never seen before, i.e. the test data. We measure the performance of the
network using the accuracy score. The accuracy is as you would expect just the
number of images correctly labeled divided by the total number of images. A
perfect classifier will have an accuracy score of 1.
Pn
I(ŷi = yi )
Accuracy = i=1 ,
n
where I is the indicator function, 1 if ŷi = yi and 0 otherwise.
epochs = 100
batch_size = 100
dnn = NeuralNetwork(X_train, Y_train_onehot, eta=eta, lmbd=lmbd, epochs=epochs, batch_size=batch_s

n_hidden_neurons=n_hidden_neurons, n_categories=n_categories)
dnn.train()
test_predict = dnn.predict(X_test)
# accuracy score from scikit library

print("Accuracy score on test set: ", accuracy_score(Y_test, test_predict))
# equivalent in numpy
def accuracy_score_numpy(Y_test, Y_pred):
return np.sum(Y_test == Y_pred) / len(Y_test)
#print("Accuracy score on test set: ", accuracy_score_numpy(Y_test, test_predict))
Adjust hyperparameters
We now perform a grid search to find the optimal hyperparameters for the
network. Note that we are only using 1 layer with 50 neurons, and human
performance is estimated to be around 98% (2% error rate).
eta_vals = np.logspace(-5, 1, 7)
lmbd_vals = np.logspace(-5, 1, 7)
# store the models for later use
DNN_numpy = np.zeros((len(eta_vals), len(lmbd_vals)), dtype=object)
# grid search
for i, eta in enumerate(eta_vals):
for j, lmbd in enumerate(lmbd_vals):
dnn = NeuralNetwork(X_train, Y_train_onehot, eta=eta, lmbd=lmbd, epochs=epochs, batch_size
n_hidden_neurons=n_hidden_neurons, n_categories=n_categories)
dnn.train()
DNN_numpy[i][j] = dnn
test_predict = dnn.predict(X_test)
30
print("Learning rate = ", eta)
print("Lambda = ", lmbd)
print("Accuracy score on test set: ", accuracy_score(Y_test, test_predict))
print()
Visualization
# visual representation of grid search
# uses seaborn heatmap, you can also do this with matplotlib imshow
import seaborn as sns
sns.set()
train_accuracy = np.zeros((len(eta_vals), len(lmbd_vals)))

test_accuracy = np.zeros((len(eta_vals), len(lmbd_vals)))
for i in range(len(eta_vals)):
for j in range(len(lmbd_vals)):
dnn = DNN_numpy[i][j]
train_pred = dnn.predict(X_train)
test_pred = dnn.predict(X_test)
train_accuracy[i][j] = accuracy_score(Y_train, train_pred)

test_accuracy[i][j] = accuracy_score(Y_test, test_pred)
fig, ax = plt.subplots(figsize = (10, 10))

sns.heatmap(train_accuracy, annot=True, ax=ax, cmap="viridis")
ax.set_title("Training Accuracy")
ax.set_ylabel("$\eta$")
ax.set_xlabel("$\lambda$")
plt.show()

sns.heatmap(test_accuracy, annot=True, ax=ax, cmap="viridis")
ax.set_title("Test Accuracy")
plt.show()
scikit-learn implementation
scikit-learn focuses more on traditional machine learning methods, such as
regression, clustering, decision trees, etc. As such, it has only two types of neural
networks: Multi Layer Perceptron outputting continuous values, MPLRegressor,
and Multi Layer Perceptron outputting labels, MLPClassifier. We will see how
simple it is to use these classes.
scikit-learn implements a few improvements from our neural network, such
as early stopping, a varying learning rate, different optimization methods, etc.
We would therefore expect a better performance overall.
from sklearn.neural_network import MLPClassifier
# store models for later use
DNN_scikit = np.zeros((len(eta_vals), len(lmbd_vals)), dtype=object)
31
dnn = MLPClassifier(hidden_layer_sizes=(n_hidden_neurons), activation=’logistic’,
alpha=lmbd, learning_rate_init=eta, max_iter=epochs)
dnn.fit(X_train, Y_train)
DNN_scikit[i][j] = dnn

print("Accuracy score on test set: ", dnn.score(X_test, Y_test))
print()
Visualization
# optional
# uses seaborn heatmap, could probably do this in matplotlib
sns.set()
dnn = DNN_scikit[i][j]
train_accuracy[i][j] = accuracy_score(Y_train, train_pred)

test_accuracy[i][j] = accuracy_score(Y_test, test_pred)

plt.show()
plt.show()
Building neural networks in Tensorflow and Keras

Now we want to build on the experience gained from our neural network imple-
mentation in NumPy and scikit-learn and use it to construct a neural network
in Tensorflow. Once we have constructed a neural network in NumPy and
32
Tensorflow, building one in Keras is really quite trivial, though the performance
may suffer.
In our previous example we used only one hidden layer, and in this we will
use two. From this it should be quite clear how to build one using an arbitrary
number of hidden layers, using data structures such as Python lists or NumPy
arrays.
Tensorflow
Tensorflow is an open source library machine learning library developed by the
Google Brain team for internal use. It was released under the Apache 2.0 open
source license in November 9, 2015.
Tensorflow is a computational framework that allows you to construct machine
learning models at different levels of abstraction, from high-level, object-oriented
APIs like Keras, down to the C++ kernels that Tensorflow is built upon. The
higher levels of abstraction are simpler to use, but less flexible, and our choice
of implementation should reflect the problems we are trying to solve.
Tensorflow uses so-called graphs to represent your computation in terms
of the dependencies between individual operations, such that you first build a
Tensorflow graph to represent your model, and then create a Tensorflow session
to run the graph.
In this guide we will analyze the same data as we did in our NumPy and
scikit-learn tutorial, gathered from the MNIST database of images. We will give
an introduction to the lower level Python Application Program Interfaces (APIs),
and see how we use them to build our graph. Then we will build (effectively) the
same graph in Keras, to see just how simple solving a machine learning problem
can be.
To install tensorflow on Unix/Linux systems, use pip as
pip install tensorflow
and/or if you use anaconda, just write (or install from the graphical user
interface)
conda install tensorflow
Collect and pre-process data

import numpy as np

np.random.seed(0)

%matplotlib inline
33

print("inputs = (n_inputs, pixel_width, pixel_height) = " + str(inputs.shape))

# flatten the image

# the value -1 means dimension is inferred from the remaining dimensions: 8x8 = 64
inputs = inputs.reshape(n_inputs, -1)
print("X = (n_inputs, n_features) = " + str(inputs.shape))


plt.axis(’off’)
plt.show()
from keras.utils import to_categorical

# one-hot representation of labels

labels = to_categorical(labels)
# split into train and test data
train_size = 0.8
Using TensorFlow backend

1. Define model and architecture
2. Choose cost function and optimizer
import tensorflow as tf
class NeuralNetworkTensorflow:
def __init__(
self,
X_train,
Y_train,
X_test,
Y_test,
n_neurons_layer1=100,
34
n_neurons_layer2=50,
n_categories=2,
epochs=10,
batch_size=100,
eta=0.1,
lmbd=0.0):
# keep track of number of steps

self.global_step = tf.Variable(0, dtype=tf.int32, trainable=False, name=’global_step’)
self.X_train = X_train
self.Y_train = Y_train
self.X_test = X_test
self.Y_test = Y_test
self.n_inputs = X_train.shape[0]
self.n_features = X_train.shape[1]
self.n_neurons_layer1 = n_neurons_layer1
self.n_neurons_layer2 = n_neurons_layer2
self.eta = eta
self.lmbd = lmbd
# build network piece by piece

# name scopes (with) are used to enforce creation of new variables
# https://www.tensorflow.org/guide/variables
self.create_placeholders()
self.create_DNN()
self.create_loss()
self.create_optimiser()
self.create_accuracy()
def create_placeholders(self):
# placeholders are fine here, but "Datasets" are the preferred method
# of streaming data into a model
with tf.name_scope(’data’):
self.X = tf.placeholder(tf.float32, shape=(None, self.n_features), name=’X_data’)
self.Y = tf.placeholder(tf.float32, shape=(None, self.n_categories), name=’Y_data’)
def create_DNN(self):
with tf.name_scope(’DNN’):
# the weights are stored to calculate regularization loss later
# Fully connected layer 1

self.W_fc1 = self.weight_variable([self.n_features, self.n_neurons_layer1], name=’fc1’
b_fc1 = self.bias_variable([self.n_neurons_layer1], name=’fc1’, dtype=tf.float32)
a_fc1 = tf.nn.sigmoid(tf.matmul(self.X, self.W_fc1) + b_fc1)
# Fully connected layer 2

self.W_fc2 = self.weight_variable([self.n_neurons_layer1, self.n_neurons_layer2], name
b_fc2 = self.bias_variable([self.n_neurons_layer2], name=’fc2’, dtype=tf.float32)
a_fc2 = tf.nn.sigmoid(tf.matmul(a_fc1, self.W_fc2) + b_fc2)
# Output layer
self.W_out = self.weight_variable([self.n_neurons_layer2, self.n_categories], name=’ou
b_out = self.bias_variable([self.n_categories], name=’out’, dtype=tf.float32)
self.z_out = tf.matmul(a_fc2, self.W_out) + b_out
35
def create_loss(self):
with tf.name_scope(’loss’):
softmax_loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits_v2(labels=self.Y
regularizer_loss_fc1 = tf.nn.l2_loss(self.W_fc1)
regularizer_loss_fc2 = tf.nn.l2_loss(self.W_fc2)
regularizer_loss_out = tf.nn.l2_loss(self.W_out)
regularizer_loss = self.lmbd*(regularizer_loss_fc1 + regularizer_loss_fc2 + regularize
self.loss = softmax_loss + regularizer_loss
def create_accuracy(self):
with tf.name_scope(’accuracy’):
probabilities = tf.nn.softmax(self.z_out)
predictions = tf.argmax(probabilities, axis=1)
labels = tf.argmax(self.Y, axis=1)
correct_predictions = tf.equal(predictions, labels)
correct_predictions = tf.cast(correct_predictions, tf.float32)
self.accuracy = tf.reduce_mean(correct_predictions)
def create_optimiser(self):
with tf.name_scope(’optimizer’):
self.optimizer = tf.train.GradientDescentOptimizer(learning_rate=self.eta).minimize(se
def weight_variable(self, shape, name=’’, dtype=tf.float32):
initial = tf.truncated_normal(shape, stddev=0.1)
return tf.Variable(initial, name=name, dtype=dtype)
def bias_variable(self, shape, name=’’, dtype=tf.float32):

initial = tf.constant(0.1, shape=shape)
def fit(self):
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
chosen_datapoints = np.random.choice(data_indices, size=self.batch_size, repla
batch_X, batch_Y = self.X_train[chosen_datapoints], self.Y_train[chosen_datapo
sess.run([DNN.loss, DNN.optimizer],
feed_dict={DNN.X: batch_X,
DNN.Y: batch_Y})
accuracy = sess.run(DNN.accuracy,
feed_dict={DNN.X: batch_X,
DNN.Y: batch_Y})
step = sess.run(DNN.global_step)
self.train_loss, self.train_accuracy = sess.run([DNN.loss, DNN.accuracy],

feed_dict={DNN.X: self.X_train,
DNN.Y: self.Y_train})
self.test_loss, self.test_accuracy = sess.run([DNN.loss, DNN.accuracy],
feed_dict={DNN.X: self.X_test,
DNN.Y: self.Y_test})
36
Optimizing and using gradient descent
epochs = 100
batch_size = 100
n_neurons_layer1 = 100
n_neurons_layer2 = 50
n_categories = 10
DNN_tf = np.zeros((len(eta_vals), len(lmbd_vals)), dtype=object)

DNN = NeuralNetworkTensorflow(X_train, Y_train, X_test, Y_test,
n_neurons_layer1, n_neurons_layer2, n_categories,
epochs=epochs, batch_size=batch_size, eta=eta, lmbd=lmbd)
DNN.fit()
DNN_tf[i][j] = DNN

print("Test accuracy: %.3f" % DNN.test_accuracy)
print()
# optional
sns.set()
DNN = DNN_tf[i][j]
train_accuracy[i][j] = DNN.train_accuracy
test_accuracy[i][j] = DNN.test_accuracy

plt.show()

plt.show()
# optional
# we can use log files to visualize our graph in Tensorboard
writer = tf.summary.FileWriter(’logs/’)
writer.add_graph(tf.get_default_graph())
37
Using Keras
Keras is a high level neural network that supports Tensorflow, CTNK and
Theano as backends. If you have Tensorflow installed Keras is available through
the tf.keras module. If you have Anaconda installed you may run the following
command
conda install keras
Alternatively, if you have Tensorflow or one of the other supported backends

install you may use the pip package manager:
pip install keras
or look up the instructions here.

from tensorflow.keras.layers import Input
from tensorflow.keras.models import Sequential #This allows appending layers to existing mode
from tensorflow.keras.layers import Dense #This allows defining the characteristics of a
from tensorflow.keras import optimizers #This allows using whichever optimiser we want
from tensorflow.keras import regularizers #This allows using whichever regularizer we wa
from tensorflow.keras.utils import to_categorical #This allows using categorical cross entropy a
def create_neural_network_keras(n_neurons_layer1, n_neurons_layer2, n_categories, eta, lmbd):

model = Sequential()
model.add(Dense(n_neurons_layer1, activation=’sigmoid’, kernel_regularizer=regularizers.l2(lmb
model.add(Dense(n_neurons_layer2, activation=’sigmoid’, kernel_regularizer=regularizers.l2(lmb
model.add(Dense(n_categories, activation=’softmax’))
sgd = optimizers.SGD(lr=eta)
model.compile(loss=’categorical_crossentropy’, optimizer=sgd, metrics=[’accuracy’])
return model
DNN_keras = np.zeros((len(eta_vals), len(lmbd_vals)), dtype=object)

DNN = create_neural_network_keras(n_neurons_layer1, n_neurons_layer2, n_categories,
eta=eta, lmbd=lmbd)
DNN.fit(X_train, Y_train, epochs=epochs, batch_size=batch_size, verbose=0)
scores = DNN.evaluate(X_test, Y_test)
DNN_keras[i][j] = DNN
print("Test accuracy: %.3f" % scores[1])
print()
# optional
sns.set()
38
DNN = DNN_keras[i][j]
train_accuracy[i][j] = DNN.evaluate(X_train, Y_train)[1]

test_accuracy[i][j] = DNN.evaluate(X_test, Y_test)[1]

plt.show()

plt.show()
The Breast Cancer Data, now with Keras

import numpy as np
from sklearn.model_selection import train_test_split as splitter
from sklearn.datasets import load_breast_cancer
import pickle
import os
"""Load breast cancer dataset"""

np.random.seed(0) #create same seed for random number every time
cancer=load_breast_cancer() #Download breast cancer dataset
inputs=cancer.data #Feature matrix of 569 rows (samples) and 30 columns (param

outputs=cancer.target #Label array of 569 rows (0 for benign and 1 for malignant)
labels=cancer.feature_names[0:30]
print(’The content of the breast cancer dataset is:’) #Print information about the datasets
print(labels)
print(’-------------------------’)
print("inputs = " + str(inputs.shape))
print("outputs = " + str(outputs.shape))
print("labels = "+ str(labels.shape))
x=inputs #Reassign the Feature and Label matrices to other variables
39
y=outputs
#%%
# Visualisation of dataset (for correlation analysis)
plt.figure()
plt.scatter(x[:,0],x[:,2],s=40,c=y,cmap=plt.cm.Spectral)
plt.xlabel(’Mean radius’,fontweight=’bold’)
plt.ylabel(’Mean perimeter’,fontweight=’bold’)
plt.show()
plt.figure()
plt.scatter(x[:,5],x[:,6],s=40,c=y, cmap=plt.cm.Spectral)
plt.xlabel(’Mean compactness’,fontweight=’bold’)
plt.ylabel(’Mean concavity’,fontweight=’bold’)
plt.show()
plt.figure()
plt.xlabel(’Mean radius’,fontweight=’bold’)
plt.ylabel(’Mean texture’,fontweight=’bold’)
plt.show()
plt.figure()
plt.xlabel(’Mean perimeter’,fontweight=’bold’)
plt.ylabel(’Mean compactness’,fontweight=’bold’)
plt.show()
# Generate training and testing datasets

#Select features relevant to classification (texture,perimeter,compactness and symmetery)
#and add to input matrix
temp1=np.reshape(x[:,1],(len(x[:,1]),1))
temp2=np.reshape(x[:,2],(len(x[:,2]),1))
X=np.hstack((temp1,temp2))
temp=np.reshape(x[:,5],(len(x[:,5]),1))
X=np.hstack((X,temp))
temp=np.reshape(x[:,8],(len(x[:,8]),1))
X=np.hstack((X,temp))
X_train,X_test,y_train,y_test=splitter(X,y,test_size=0.1) #Split datasets into training and test
y_train=to_categorical(y_train) #Convert labels to categorical when using categorical cross en

y_test=to_categorical(y_test)
del temp1,temp2,temp
# %%
# Define tunable parameters"
eta=np.logspace(-3,-1,3) #Define vector of learning rates (parameter to SGD opt

lamda=0.01 #Define hyperparameter
n_layers=2 #Define number of hidden layers in the model
n_neuron=np.logspace(0,3,4,dtype=int) #Define number of neurons per layer
epochs=100 #Number of reiterations over the input data
40
batch_size=100 #Number of samples per gradient update
# %%
"""Define function to return Deep Neural Network model"""
def NN_model(inputsize,n_layers,n_neuron,eta,lamda):
model=Sequential()
for i in range(n_layers): #Run loop to add hidden layers to the model
if (i==0): #First layer requires input dimensions
model.add(Dense(n_neuron,activation=’relu’,kernel_regularizer=regularizers.l2(lamda),i
else: #Subsequent layers are capable of automatic shape inferencing
model.add(Dense(n_neuron,activation=’relu’,kernel_regularizer=regularizers.l2(lamda)))
model.add(Dense(2,activation=’softmax’)) #2 outputs - ordered and disordered (softmax for pro
sgd=optimizers.SGD(lr=eta)
model.compile(loss=’categorical_crossentropy’,optimizer=sgd,metrics=[’accuracy’])
return model
Train_accuracy=np.zeros((len(n_neuron),len(eta))) #Define matrices to store accuracy scores a

Test_accuracy=np.zeros((len(n_neuron),len(eta))) #of learning rate and number of hidden neur
for i in range(len(n_neuron)): #run loops over hidden neurons and learning rates to calculate
for j in range(len(eta)): #accuracy scores
DNN_model=NN_model(X_train.shape[1],n_layers,n_neuron[i],eta[j],lamda)
DNN_model.fit(X_train,y_train,epochs=epochs,batch_size=batch_size,verbose=1)
Train_accuracy[i,j]=DNN_model.evaluate(X_train,y_train)[1]
Test_accuracy[i,j]=DNN_model.evaluate(X_test,y_test)[1]
def plot_data(x,y,data,title=None):
# plot results
fontsize=16
fig = plt.figure()
cax = ax.matshow(data, interpolation=’nearest’, vmin=0, vmax=1)
cbar=fig.colorbar(cax)
cbar.ax.set_ylabel(’accuracy (%)’,rotation=90,fontsize=fontsize)
cbar.set_ticks([0,.2,.4,0.6,0.8,1.0])
cbar.set_ticklabels([’0%’,’20%’,’40%’,’60%’,’80%’,’100%’])
# put text on matrix elements
for i, x_val in enumerate(np.arange(len(x))):
for j, y_val in enumerate(np.arange(len(y))):
c = "${0:.1f}\\%$".format( 100*data[j,i])
ax.text(x_val, y_val, c, va=’center’, ha=’center’)
# convert axis vaues to to string labels

x=[str(i) for i in x]
y=[str(i) for i in y]
ax.set_xticklabels([’’]+x)
ax.set_yticklabels([’’]+y)
ax.set_xlabel(’$\\mathrm{learning\\ rate}$’,fontsize=fontsize)
ax.set_ylabel(’$\\mathrm{hidden\\ neurons}$’,fontsize=fontsize)
41
if title is not None:
ax.set_title(title)
plt.tight_layout()
plt.show()
plot_data(eta,n_neuron,Train_accuracy, ’training’)
plot_data(eta,n_neuron,Test_accuracy, ’testing’)
Which activation function should I use?

The Back propagation algorithm we derived above works by going from the
output layer to the input layer, propagating the error gradient on the way. Once
the algorithm has computed the gradient of the cost function with regards to
each parameter in the network, it uses these gradients to update each parameter
with a Gradient Descent (GD) step.
Unfortunately for us, the gradients often get smaller and smaller as the
algorithm progresses down to the first hidden layers. As a result, the GD update
leaves the lower layer connection weights virtually unchanged, and training never
converges to a good solution. This is known in the literature as the vanishing
gradients problem.
In other cases, the opposite can happen, namely the the gradients can grow
bigger and bigger. The result is that many of the layers get large updates of the
weights the algorithm diverges. This is the exploding gradients problem,
which is mostly encountered in recurrent neural networks. More generally, deep
neural networks suffer from unstable gradients, different layers may learn at
widely different speeds
Is the Logistic activation function (Sigmoid) our choice?

Although this unfortunate behavior has been empirically observed for quite a
while (it was one of the reasons why deep neural networks were mostly abandoned
for a long time), it is only around 2010 that significant progress was made in
understanding it.
A paper titled Understanding the Difficulty of Training Deep Feedforward
Neural Networks by Xavier Glorot and Yoshua Bengio found that the problems
with the popular logistic sigmoid activation function and the weight initialization
technique that was most popular at the time, namely random initialization using
a normal distribution with a mean of 0 and a standard deviation of 1.
They showed that with this activation function and this initialization scheme,
the variance of the outputs of each layer is much greater than the variance of its
inputs. Going forward in the network, the variance keeps increasing after each
layer until the activation function saturates at the top layers. This is actually
made worse by the fact that the logistic function has a mean of 0.5, not 0 (the
hyperbolic tangent function has a mean of 0 and behaves slightly better than
the logistic function in deep networks).
42
The derivative of the Logistic funtion
Looking at the logistic activation function, when inputs become large (negative
or positive), the function saturates at 0 or 1, with a derivative extremely close to
0. Thus when backpropagation kicks in, it has virtually no gradient to propagate
back through the network, and what little gradient exists keeps getting diluted
as backpropagation progresses down through the top layers, so there is really
nothing left for the lower layers.
In their paper, Glorot and Bengio propose a way to significantly alleviate
this problem. We need the signal to flow properly in both directions: in the
forward direction when making predictions, and in the reverse direction when
backpropagating gradients. We don’t want the signal to die out, nor do we want
it to explode and saturate. For the signal to flow properly, the authors argue
that we need the variance of the outputs of each layer to be equal to the variance
of its inputs, and we also need the gradients to have equal variance before and
after flowing through a layer in the reverse direction.
One of the insights in the 2010 paper by Glorot and Bengio was that the
vanishing/exploding gradients problems were in part due to a poor choice of
activation function. Until then most people had assumed that if Nature had
chosen to use roughly sigmoid activation functions in biological neurons, they
must be an excellent choice. But it turns out that other activation functions
behave much better in deep neural networks, in particular the ReLU activation
function, mostly because it does not saturate for positive values (and also because
it is quite fast to compute).
The RELU function family

The ReLU activation function suffers from a problem known as the dying ReLUs:
during training, some neurons effectively die, meaning they stop outputting
anything other than 0.
In some cases, you may find that half of your network’s neurons are dead,
especially if you used a large learning rate. During training, if a neuron’s weights
get updated such that the weighted sum of the neuron’s inputs is negative, it
will start outputting 0. When this happen, the neuron is unlikely to come back
to life since the gradient of the ReLU function is 0 when its input is negative.
To solve this problem, nowadays practitioners use a variant of the ReLU
function, such as the leaky ReLU discussed above or the so-called exponential
linear unit (ELU) function

α (exp (z) − 1) z < 0,
ELU (z) =
z z ≥ 0.
Which activation function should we use?

In general it seems that the ELU activation function is better than the leaky
ReLU function (and its variants), which is better than ReLU. ReLU performs
better than tanh which in turn performs better than the logistic function.
43
If runtime performance is an issue, then you may opt for the leaky ReLU
function over the ELU function If you don’t want to tweak yet another hyperpa-
rameter, you may just use the default α of 0.01 for the leaky ReLU, and 1 for
ELU. If you have spare time and computing power, you can use cross-validation
or bootstrap to evaluate other activation functions.
A top-down perspective on Neural networks

The first thing we would like to do is divide the data into two or three parts. A
training set, a validation or dev (development) set, and a test set. The test set
is the data on which we want to make predictions. The dev set is a subset of the
training data we use to check how well we are doing out-of-sample, after training
the model on the training dataset. We use the validation error as a proxy for the
test error in order to make tweaks to our model. It is crucial that we do not use
any of the test data to train the algorithm. This is a cardinal sin in ML. Then:
• Estimate optimal error rate

• Minimize underfitting (bias) on training data set.
• Make sure you are not overfitting.
If the validation and test sets are drawn from the same distributions, then a good
performance on the validation set should lead to similarly good performance on
the test set.
However, sometimes the training data and test data differ in subtle ways
because, for example, they are collected using slightly different methods, or
because it is cheaper to collect data in one way versus another. In this case,
there can be a mismatch between the training and test data. This can lead to the
neural network overfitting these small differences between the test and training
sets, and a poor performance on the test set despite having a good performance
on the validation set. To rectify this, Andrew Ng suggests making two validation
or dev sets, one constructed from the training data and one constructed from
the test data. The difference between the performance of the algorithm on these
two validation sets quantifies the train-test mismatch. This can serve as another
important diagnostic when using DNNs for supervised learning.
Limitations of supervised learning with deep networks

Like all statistical methods, supervised learning using neural networks has
important limitations. This is especially important when one seeks to apply
these methods, especially to physics problems. Like all tools, DNNs are not
a universal solution. Often, the same or better performance on a task can be
achieved by using a few hand-engineered features (or even a collection of random
features).
Here we list some of the important limitations of supervised neural network
based models.
44
• Need labeled data. All supervised learning methods, DNNs for super-
vised learning require labeled data. Often, labeled data is harder to acquire
than unlabeled data (e.g. one must pay for human experts to label images).
• Supervised neural networks are extremely data intensive. DNNs
are data hungry. They perform best when data is plentiful. This is doubly
so for supervised methods where the data must also be labeled. The utility
of DNNs is extremely limited if data is hard to acquire or the datasets are
small (hundreds to a few thousand samples). In this case, the performance
of other methods that utilize hand-engineered features can exceed that of
DNNs.
• Homogeneous data. Almost all DNNs deal with homogeneous data of

one type. It is very hard to design architectures that mix and match data
types (i.e. some continuous variables, some discrete variables, some time
series). In applications beyond images, video, and language, this is often
what is required. In contrast, ensemble models like random forests or
gradient-boosted trees have no difficulty handling mixed data types.
• Many problems are not about prediction. In natural science we are

often interested in learning something about the underlying distribution
that generates the data. In this case, it is often difficult to cast these
ideas in a supervised learning setting. While the problems are related, it is
possible to make good predictions with a wrong model. The model might
or might not be useful for understanding the underlying science.
Some of these remarks are particular to DNNs, others are shared by all supervised
learning methods. This motivates the use of unsupervised methods which in
part circumvent these problems.
Examples: Pulsar identification

Pulsar classification can be one of the best training grounds for performing
machine learning in astrophysics. These pulsars are said to be pulsating radio
sources, which have now been identified to be caused by rapidly rotating highly
magnetized neutron stars and are detectable here on Earth . One of the charac-
teristic properties of these pulsars is that they exhibit periodic bursts of emission
produced by their radio emitting jets. The direction of their emission also
rotates with them and sweep the sky like a lighthouse. This gives astronomers
information about the phenomenon as they observe a pulse of radio emission
each time one of the jets points towards the Earth.
The study and detection of pulsars provide a wealth of information about the
physics of neutron stars. They are also used as probes of stellar evolution. In
addition, they are being used to test or verify some concepts in general relativity
due to their extremely high densities. These allowed them to be good observables
in detecting and mapping gravitational wave signatures. One problem, however,
is that they are very difficult to identify in the large stream of data from radio
45
telescopes. Moreover, a lot of man-made sources of radio frequency interference
are also present, which can produce the same signals as pulsars. Hence, the
classification of pulsars from possible candidate data is of great importance.
In reality, pulsars are very weak radio sources. They are classified from a
given data sample by first extracting information from the pool of data and then
identifying which features or characteristics are relevant. Since the individual
pulses are very much different, astronomers particularly stack them up and
generate an integrated pulse profile for pulsar classification. This profile is
the coherent addition of thousands of pulses together in a process known as
folding. Moreover, pulses will arrive at various times across different radio
frequencies. The delay of these pulses from frequency to frequency is called
dispersion and is said to be caused by the ionized inter-stellar medium. The
method usually employed by astronomers is that they fit for the shape of the delay
as to reduce its negative effect. However, as with all kinds of fitting procedures,
there would always be an uncertainty associated with it. This is expressed in
the so-called DM-SNR (dispersion-measure-signal-to-noise-ratio) curve. Both
the integrated profile curve and DM-SNR curve are used in identifying possible
pulsar candidates. These curves provide eight numerical characteristic features
as depicted and listed below.
Pulsar data set

For this dataset, there is already an initial classification of the potential pulsar
candidates as pulsars and non-pulsars by the astronomy community. We aim
here to perform machine learning and try to build a model that can detect
patterns within the data. This will eventually lead to the correct classification
of new potential pulsars that will soon be observed.
# reading and handling the data
import pandas as pd
import warnings
warnings.filterwarnings("ignore")
data = pd.read_csv("pulsar_stars.csv")
# print a part of the dataset
data.head()
print ("Number of rows :",data.shape[0])
print ("Number of columns :",data.shape[1])
print ("data info :",data.info())
print (data.isnull().sum())
# import packages for plotting
plt.figure(figsize=(12,6))
plt.subplot(121)
ax = sns.countplot(y = data["target_class"],
palette=["r","g"],
linewidth=1,
edgecolor="k"*2)
for i,j in enumerate(data["target_class"].value_counts().values):
ax.text(.7,i,j,weight = "bold",fontsize = 27)
46
plt.title("Count for target variable in dataset")
plt.subplot(122)
plt.pie(data["target_class"].value_counts().values,
labels=["not pulsar stars","pulsar stars"],
autopct="%1.0f%%",wedgeprops={"linewidth":2,"edgecolor":"white"})
my_circ = plt.Circle((0,0),.7,color = "white")
plt.gca().add_artist(my_circ)
plt.subplots_adjust(wspace = .2)
plt.title("Proportion of target variable in dataset")
plt.show()
Number of pulsar candidates

The dataset contains 17898 candidates for which only 1639 (≈ 9%) are classified
as real pulsars. In addition, the dataset consists of features rather than raw
data from observations. The eight features considered in the study are given
above as column headers. The first four are the usual statistics obtained from
the integrated pulse profile. A class label (targetclass) is also included which
determines if the candidate is considered to be a pulsar (1) or not (0). This
serves as our target column for the analysis.
Preprocessing and Statistical Analysis of the Data. When dealing with

all sorts of data, an important first step is to often get a sense on how the
variables are distributed. As seen above, the dataset can be considered clean and
complete as there are no missing data and the number of features is relatively
small. This reduces extra steps towards feature engineering as there is not
much to do for artificial design features. Thus, we can see readily that the
preprocessing stage only involves two main issues. First is the possible problem
of overfitting since the difference between the numbers of candidates identified
as pulsars and non-pulsars is large, with a ratio of about 1:10. The other issue
is the huge distribution for the range differences of the eight features. This is
evident in the figures here. The plots were done using the distplot() function
in the package seaborn. This draws a histogram and fit a kernel density estimate
(KDE).
# import package
import itertools
# rename column headers for convenience
data = data.rename(columns={" Mean of the integrated profile": "Mean profile", " Standard deviatio
" Excess kurtosis of the integrated profile": "Kurtosis profile", " Skewness o
" Mean of the DM-SNR curve": "Mean DM-SNR", " Standard deviation of the DM-SNR
" Excess kurtosis of the DM-SNR curve": "Kurtosis DM-SNR", " Skewness of the D
"target_class": "Target class"})
# distribution of variable in the dataset

columns = [’Mean profile’, ’StD profile’, ’Kurtosis profile’, ’Skewness profile’,
’Mean DM-SNR’, ’StD DM-SNR’, ’Kurtosis DM-SNR’,
’Skewness DM-SNR’]
length = len(columns)
colors = ["r","g","b","m","y","c","k","orange"]
47
for i,j,k in itertools.zip_longest(columns,range(length),colors):
plt.subplot(length/2,length/4,j+1)
sns.distplot(data[i],color=k)
plt.title(i)
plt.subplots_adjust(hspace = .3)
plt.axvline(data[i].mean(),color = "k",linestyle="dashed",label="MEAN")
plt.axvline(data[i].std(),color = "b",linestyle="dotted",label="STANDARD DEVIATION")
plt.legend(loc="upper right")
Before delving into any analysis, we would like to decrease the amount of
data to look at. This can be achieved by performing a correlation plot. As
seen in Figure 5, the features that correlate the most with the class (pulsar or
non-pulsar) are the mean, the excess kurtosis, and the skewness of the integrated
profile. We could choose to only analyze these, but this would mean that pulsars
can be identified with only one of the two plots that astrophysicists use to classify
the data. In this case, colinearity may become a problem since these features
are strongly correlated with each other.
# correlation matrix
corr_matrix = data.corr()
# Pair plots
sns.pairplot(data,hue = "Target class")
plt.title("Pair plot for variables")
# plt.savefig(’pairplot.png’,bbox_inches=’tight’)
plt.show()
# Violin Plot
columns = [x for x in data.columns if x not in ["Target class"]]
length = len(columns)
for i,j in itertools.zip_longest(columns,range(length)):
plt.subplot(length/2,length/4,j+1)
sns.violinplot(x=data["Target class"],y=data[i],
palette=["Orangered","lime"],alpha=.5)
plt.title(i)
#plt.savefig(’violinplot.png’,bbox_inches=’tight’)
plt.show()
# Logistc regression
import numpy as np
from sklearn.linear_model import LogisticRegression
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import KFold
from sklearn.metrics import confusion_matrix
# We define the hyperparameters to be used
nlambdas = 500 # why not? the code runs relatively fast
lmbdas = np.logspace(-5, 5, nlambdas)
kfold = KFold(n_splits = 5) #cross validation spliting
# We preallocate data
# true values that will be found later
48
train_accuracy=np.zeros(lmbdas.shape,np.float64)
test_accuracy=np.zeros(lmbdas.shape,np.float64)
train_red_accuracy=np.zeros(lmbdas.shape,np.float64)
test_red_accuracy=np.zeros(lmbdas.shape,np.float64)
# dummy arrays to be averaged later on
train_accuracy_d=np.zeros(5,np.float64)
test_accuracy_d=np.zeros(5,np.float64)
train_red_accuracy_d=np.zeros(5,np.float64)
test_red_accuracy_d=np.zeros(5,np.float64)
# We create the design matrix X and separate the labels into Y
x_fea = [x for x in data.columns if x not in [’Target class’]]
#print(x_fea)
X = np.zeros((data.shape[0],data.shape[1]-1))
X_red = np.zeros((data.shape[0],3))
Y = np.zeros(data.shape[0])
for i,feature in enumerate(x_fea): # Here we just take the variables of interest
X[:,i] = data[feature]
if ’Mean profile’ == feature:
X_red[:,0]
if ’Kurtosis profile’ == feature:
X_red[:,1]
if ’Skewness profile’== feature:
X_red[:,2]
Y[:] = data[’Target class’]
# We perform a logistic regression for each value of lambda
for i,lmbda in enumerate(lmbdas):
#define model
logreg = LogisticRegression(C=1.0/lmbda,solver=’liblinear’)
# Perform the cross-validation

j = 0
for train_inds, test_inds in kfold.split(X):
# Do the split
X_train = X[train_inds]
X_red_train = X_red[train_inds]
Y_train = Y[train_inds]
X_test = X[test_inds]
X_red_test = X_red[test_inds]
Y_test = Y[test_inds]
# We will scale the data
scaler = StandardScaler()
scaler.fit(X_train)
# first on full data
X_train = scaler.transform(X_train)
X_test = scaler.transform(X_test)
# then rescale and do on reduced data
scaler.fit(X_red_train)
X_red_train = scaler.transform(X_red_train)
X_red_test = scaler.transform(X_red_test)
del scaler
# calculate accuracies for the k fold

logreg.fit(X_train, Y_train)
train_accuracy_d[j]=logreg.score(X_train,Y_train)
test_accuracy_d[j]=logreg.score(X_test,Y_test)
logreg.fit(X_red_train, Y_train)
train_red_accuracy_d[j]=logreg.score(X_red_train,Y_train)
49
test_red_accuracy_d[j]=logreg.score(X_red_test,Y_test)
j += 1
del X_red_train,X_red_test,X_train,Y_train,X_test,Y_test # delete useless data
#Average to get accuracy values
train_accuracy[i]=np.mean(train_accuracy_d)
test_accuracy[i]=np.mean(test_accuracy_d)
train_red_accuracy[i]=np.mean(train_red_accuracy_d)
test_red_accuracy[i]=np.mean(test_red_accuracy_d)
plt.semilogx(lmbdas,train_accuracy,label=’train’)
plt.semilogx(lmbdas,test_accuracy,label=’test’)
plt.semilogx(lmbdas,train_red_accuracy,label=’train reduced’)
plt.semilogx(lmbdas,test_red_accuracy,’--’,label=’test reduced’)
plt.xlabel(’$\\lambda$’)
plt.ylabel(’$\\mathrm{accuracy}$’)
plt.grid()
plt.legend()
plt.show()
The neural network

# Neural Network
import warnings
import itertools
from PIL import Image
from sklearn.metrics import confusion_matrix,accuracy_score
from sklearn.neural_network import MLPClassifier
warnings.filterwarnings("ignore")
#Split data with 20% of test data:

np.random.seed(2018)
X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size = 0.2, random_state = 66)
# Define the learning rate, hyperparameter using NUMPY

n_hidden_neurons = 50
epochs = 100
# Use scikit learn for neural network
DNN_scikit = np.zeros((len(eta_vals), len(lmbd_vals)), dtype=object)
dnn = MLPClassifier(hidden_layer_sizes=(n_hidden_neurons), activation=’logistic’,
alpha=lmbd, learning_rate_init=eta, max_iter=epochs, solver=’adam’)
dnn.fit(X_train, y_train)
DNN_scikit[i][j] = dnn
# just uncomment below to print the accuracy scores
#print("Learning rate = ", eta)
#print("Lambda = ", lmbd)
#print("Accuracy score on test set: ", dnn.score(X_test, y_test))
#print()
#Plot the accuracy as function of learning rate and hyperparameter

# just uncomment the lines below to generate the plots (ctrl + /)
50
sns.set()
dnn = DNN_scikit[i][j]
train_accuracy[i][j] = accuracy_score(y_train, train_pred)
test_accuracy[i][j] = accuracy_score(y_test, test_pred)

sns.heatmap(train_accuracy, annot=True,annot_kws={"size": 18}, ax=ax, cmap="viridis")
ax.set_title("Training Accuracy",fontsize=18)
ax.set_ylabel("$\eta$",fontsize=18)
ax.set_yticklabels(eta_vals)
ax.set_xlabel("$\lambda$",fontsize=18)
ax.set_xticklabels(lmbd_vals)
plt.tick_params(labelsize=18)
sns.heatmap(test_accuracy, annot=True,annot_kws={"size": 18}, ax=ax, cmap="viridis")
ax.set_title("Test Accuracy",fontsize=18)
ax.set_ylabel("$\eta$",fontsize=18)
ax.set_yticklabels(eta_vals)
ax.set_xlabel("$\lambda$",fontsize=18)
ax.set_xticklabels(lmbd_vals)
plt.show()
#Plot confusion matrix at optimal values of learning rate and hyperameter

# just uncomment the lines below to generate the plots (ctrl + /)
dnn = MLPClassifier(hidden_layer_sizes=(n_hidden_neurons), activation=’logistic’,alpha=0.001, lear
dnn.fit(X_train,y_train)
y_pred=dnn.predict(X_test)
fig1, ax = plt.subplots(figsize = (13,10))
sns.heatmap(confusion_matrix(y_test,y_pred),annot=True,fmt = "d",linecolor="k",linewidths=3)
ax.set_xlabel(’True label’,fontsize=18)
ax.set_ylabel(’Predicted label’,fontsize=18)
ax.set_title("CONFUSION MATRIX",fontsize=20)
plt.show()
# Feature importance -->weights
coef=dnn.coefs_[0]
print (coef)
Convolutional Neural Networks (recognizing images)

Convolutional neural networks (CNNs) were developed during the last decade
of the previous century, with a focus on character recognition tasks. Nowadays,
CNNs are a central element in the spectacular success of dee learning methods.
The success in for example image classifications have made them a central tool
for most machine learning practitioners.
CNNs are very similar to ordinary Neural Networks. They are made up
of neurons that have learnable weights and biases. Each neuron receives some
51
inputs, performs a dot product and optionally follows it with a non-linearity.
The whole network still expresses a single differentiable score function: from the
raw image pixels on one end to class scores at the other. And they still have
a loss function (for example Softmax) on the last (fully-connected) layer and
all the tips/tricks we developed for learning regular Neural Networks still apply
(back propagation, gradient descent etc etc).
What is the difference? CNN architectures make the explicit assump-
tion that the inputs are images, which allows us to encode certain
properties into the architecture. These then make the forward func-
tion more efficient to implement and vastly reduce the amount of
parameters in the network.
Here we provide only a superficial overview, for the more interested, we
recommend highly the course IN5400 – Machine Learning for Image Analysis
and the slides of CS231.
Another good read is the article here https://arxiv.org/pdf/1603.07285.
pdf.
Regular NNs don’t scale well to full images

As an example, consider an image of size 32 × 32 × 3 (32 wide, 32 high, 3 color
channels), so a single fully-connected neuron in a first hidden layer of a regular
Neural Network would have 32 × 32 × 3 = 3072 weights. This amount still seems
manageable, but clearly this fully-connected structure does not scale to larger
images. For example, an image of more respectable size, say 200 × 200 × 3, would
lead to neurons that have 200 × 200 × 3 = 120, 000 weights.
We could have several such neurons, and the parameters would add up
quickly! Clearly, this full connectivity is wasteful and the huge number of
parameters would quickly lead to possible overfitting.
Figure 1: A regular 3-layer Neural Network.
3D volumes of neurons
Convolutional Neural Networks take advantage of the fact that the input consists
of images and they constrain the architecture in a more sensible way.
52
In particular, unlike a regular Neural Network, the layers of a CNN have
neurons arranged in 3 dimensions: width, height, depth. (Note that the word
depth here refers to the third dimension of an activation volume, not to the
depth of a full Neural Network, which can refer to the total number of layers in
a network.)
To understand it better, the above example of an image with an input volume
of activations has dimensions 32 × 32 × 3 (width, height, depth respectively).
The neurons in a layer will only be connected to a small region of the layer
before it, instead of all of the neurons in a fully-connected manner. Moreover,
the final output layer could for this specific image have dimensions 1 × 1 × 10,
because by the end of the CNN architecture we will reduce the full image into a
single vector of class scores, arranged along the depth dimension.
Figure 2: A CNN arranges its neurons in three dimensions (width, height,

depth), as visualized in one of the layers. Every layer of a CNN transforms
the 3D input volume to a 3D output volume of neuron activations. In this
example, the red input layer holds the image, so its width and height would
be the dimensions of the image, and the depth would be 3 (Red, Green, Blue
channels).
Layers used to build CNNs

A simple CNN is a sequence of layers, and every layer of a CNN transforms one
volume of activations to another through a differentiable function. We use three
main types of layers to build CNN architectures: Convolutional Layer, Pooling
Layer, and Fully-Connected Layer (exactly as seen in regular Neural Networks).
We will stack these layers to form a full CNN architecture.
A simple CNN for image classification could have the architecture:
• INPUT (32 × 32 × 3) will hold the raw pixel values of the image, in this
case an image of width 32, height 32, and with three color channels R,G,B.
• CONV (convolutional )layer will compute the output of neurons that are
connected to local regions in the input, each computing a dot product
between their weights and a small region they are connected to in the input
volume. This may result in volume such as [32 × 32 × 12] if we decided to
use 12 filters.
53
• RELU layer will apply an elementwise activation function, such as the
max(0, x) thresholding at zero. This leaves the size of the volume un-
changed ([32 × 32 × 12]).
• POOL (pooling) layer will perform a downsampling operation along the
spatial dimensions (width, height), resulting in volume such as [16×16×12].
• FC (i.e. fully-connected) layer will compute the class scores, resulting in
volume of size [1 × 1 × 10], where each of the 10 numbers correspond to
a class score, such as among the 10 categories of the MNIST images we
considered above . As with ordinary Neural Networks and as the name
implies, each neuron in this layer will be connected to all the numbers in
the previous volume.
Transforming images
CNNs transform the original image layer by layer from the original pixel values
to the final class scores.
Observe that some layers contain parameters and other don’t. In particular,
the CNN layers perform transformations that are a function of not only the
activations in the input volume, but also of the parameters (the weights and biases
of the neurons). On the other hand, the RELU/POOL layers will implement
a fixed function. The parameters in the CONV/FC layers will be trained with
gradient descent so that the class scores that the CNN computes are consistent
with the labels in the training set for each image.
CNNs in brief
In summary:
• A CNN architecture is in the simplest case a list of Layers that transform
the image volume into an output volume (e.g. holding the class scores)
• There are a few distinct types of Layers (e.g. CONV/FC/RELU/POOL
are by far the most popular)
• Each Layer accepts an input 3D volume and transforms it to an output
3D volume through a differentiable function
• Each Layer may or may not have parameters (e.g. CONV/FC do, RELU/POOL
don’t)
• Each Layer may or may not have additional hyperparameters (e.g. CONV/FC/POOL
do, RELU doesn’t)
For more material on convolutional networks, we strongly recommend the course
IN5400 – Machine Learning for Image Analysis and the slides of CS231 which is
taught at Stanford University (consistently ranked as one of the top computer
science programs in the world). Michael Nielsen’s book is a must read, in
particular chapter 6 which deals with CNNs.
54
CNNs in more detail, building convolutional neural net-
works in Tensorflow and Keras
As discussed above, CNNs are neural networks built from the assumption that
the inputs to the network are 2D images. This is important because the number
of features or pixels in images grows very fast with the image size, and an
enormous number of weights and biases are needed in order to build an accurate
network.
As before, we still have our input, a hidden layer and an output. What’s
novel about convolutional networks are the convolutional and pooling layers
stacked in pairs between the input and the hidden layer. In addition, the data is
no longer represented as a 2D feature matrix, instead each input is a number of
2D matrices, typically 1 for each color dimension (Red, Green, Blue).
Setting it up
It means that to represent the entire dataset of images, we require a 4D matrix
or tensor. This tensor has the dimensions:
(ninputs , npixels,width , npixels,height , depth).
The MNIST dataset again

The MNIST dataset consists of grayscale images with a pixel size of 28 × 28,
meaning we require 28 × 28 = 724 weights to each neuron in the first hidden
layer.
If we were to analyze images of size 128 × 128 we would require 128 × 128 =
16384 weights to each neuron. Even worse if we were dealing with color images,
as most images are, we have an image matrix of size 128 × 128 for each color
dimension (Red, Green, Blue), meaning 3 times the number of weights = 49152
are required for every single neuron in the first hidden layer.
Strong correlations
Images typically have strong local correlations, meaning that a small part of
the image varies little from its neighboring regions. If for example we have an
image of a blue car, we can roughly assume that a small blue part of the image
is surrounded by other blue regions.
Therefore, instead of connecting every single pixel to a neuron in the first
hidden layer, as we have previously done with deep neural networks, we can
instead connect each neuron to a small part of the image (in all 3 RGB depth
dimensions). The size of each small area is fixed, and known as a receptive.
Layers of a CNN
The layers of a convolutional neural network arrange neurons in 3D: width,
height and depth. The input image is typically a square matrix of depth 3.
55
A convolution is performed on the image which outputs a 3D volume of
neurons. The weights to the input are arranged in a number of 2D matrices,
known as filters.
Each filter slides along the input image, taking the dot product between each
small part of the image and the filter, in all depth dimensions. This is then
passed through a non-linear function, typically the Rectified Linear (ReLu)
function, which serves as the activation of the neurons in the first convolutional
layer. This is further passed through a pooling layer, which reduces the size
of the convolutional layer, e.g. by taking the maximum or average across some
small regions, and this serves as input to the next convolutional layer.
Systematic reduction
By systematically reducing the size of the input volume, through convolution and
pooling, the network should create representations of small parts of the input,
and then from them assemble representations of larger areas. The final pooling
layer is flattened to serve as input to a hidden layer, such that each neuron in the
final pooling layer is connected to every single neuron in the hidden layer. This
then serves as input to the output layer, e.g. a softmax output for classification.
Prerequisites: Collect and pre-process data

import numpy as np

np.random.seed(0)

%matplotlib inline


# RGB images have a depth of 3

# our images are grayscale so they should have a depth of 1
inputs = inputs[:,:,:,np.newaxis]
print("inputs = (n_inputs, pixel_width, pixel_height, depth) = " + str(inputs.shape))


56
plt.axis(’off’)
plt.show()
Importing Keras and Tensorflow

# representation of labels
labels = to_categorical(labels)
# split into train and test data
# one-liner from scikit-learn library
train_size = 0.8
Using TensorFlow backend

We need to define model and architecture and choose cost function and optmizer.
class ConvolutionalNeuralNetworkTensorflow:
def __init__(
self,
X_train,
Y_train,
X_test,
Y_test,
n_filters=10,
n_neurons_connected=50,
n_categories=10,
receptive_field=3,
stride=1,
padding=1,
epochs=10,
batch_size=100,
eta=0.1,
lmbd=0.0):
self.global_step = tf.Variable(0, dtype=tf.int32, trainable=False, name=’global_step’)
self.X_train = X_train
self.Y_train = Y_train
self.X_test = X_test
57
self.Y_test = Y_test
self.n_inputs, self.input_width, self.input_height, self.depth = X_train.shape
self.n_filters = n_filters
self.n_downsampled = int(self.input_width*self.input_height*n_filters / 4)
self.n_neurons_connected = n_neurons_connected
self.receptive_field = receptive_field
self.stride = stride
self.strides = [stride, stride, stride, stride]
self.padding = padding
self.eta = eta
self.lmbd = lmbd
self.create_placeholders()
self.create_CNN()
self.create_loss()
self.create_optimiser()
self.create_accuracy()
def create_placeholders(self):
with tf.name_scope(’data’):
self.X = tf.placeholder(tf.float32, shape=(None, self.input_width, self.input_height, self
self.Y = tf.placeholder(tf.float32, shape=(None, self.n_categories), name=’Y_data’)
def create_CNN(self):
with tf.name_scope(’CNN’):
# Convolutional layer
self.W_conv = self.weight_variable([self.receptive_field, self.receptive_field, self.depth
b_conv = self.weight_variable([self.n_filters], name=’conv’, dtype=tf.float32)
z_conv = tf.nn.conv2d(self.X, self.W_conv, self.strides, padding=’SAME’, name=’conv’) + b_
a_conv = tf.nn.relu(z_conv)
# 2x2 max pooling
a_pool = tf.nn.max_pool(a_conv, [1, 2, 2, 1], [1, 2, 2, 1], padding=’SAME’, name=’pool’)
# Fully connected layer
a_pool_flat = tf.reshape(a_pool, [-1, self.n_downsampled])
self.W_fc = self.weight_variable([self.n_downsampled, self.n_neurons_connected], name=’fc’
b_fc = self.bias_variable([self.n_neurons_connected], name=’fc’, dtype=tf.float32)
a_fc = tf.nn.relu(tf.matmul(a_pool_flat, self.W_fc) + b_fc)
# Output layer
self.W_out = self.weight_variable([self.n_neurons_connected, self.n_categories], name=’out
b_out = self.bias_variable([self.n_categories], name=’out’, dtype=tf.float32)
self.z_out = tf.matmul(a_fc, self.W_out) + b_out
def create_loss(self):
with tf.name_scope(’loss’):
softmax_loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits_v2(labels=self.Y, lo
regularizer_loss_conv = tf.nn.l2_loss(self.W_conv)
regularizer_loss_fc = tf.nn.l2_loss(self.W_fc)
regularizer_loss_out = tf.nn.l2_loss(self.W_out)
regularizer_loss = self.lmbd*(regularizer_loss_conv + regularizer_loss_fc + regularizer_lo
self.loss = softmax_loss + regularizer_loss
58
def create_accuracy(self):
with tf.name_scope(’accuracy’):
probabilities = tf.nn.softmax(self.z_out)
predictions = tf.argmax(probabilities, 1)
labels = tf.argmax(self.Y, 1)
correct_predictions = tf.equal(predictions, labels)
correct_predictions = tf.cast(correct_predictions, tf.float32)
self.accuracy = tf.reduce_mean(correct_predictions)
def create_optimiser(self):
with tf.name_scope(’optimizer’):
self.optimizer = tf.train.GradientDescentOptimizer(learning_rate=self.eta).minimize(self.l
def weight_variable(self, shape, name=’’, dtype=tf.float32):
initial = tf.truncated_normal(shape, stddev=0.1)
def bias_variable(self, shape, name=’’, dtype=tf.float32):
initial = tf.constant(0.1, shape=shape)
def fit(self):
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
chosen_datapoints = np.random.choice(data_indices, size=self.batch_size, replace=F
batch_X, batch_Y = self.X_train[chosen_datapoints], self.Y_train[chosen_datapoints
sess.run([CNN.loss, CNN.optimizer],
feed_dict={CNN.X: batch_X,
CNN.Y: batch_Y})
accuracy = sess.run(CNN.accuracy,
feed_dict={CNN.X: batch_X,
CNN.Y: batch_Y})
step = sess.run(CNN.global_step)
self.train_loss, self.train_accuracy = sess.run([CNN.loss, CNN.accuracy],
feed_dict={CNN.X: self.X_train,
CNN.Y: self.Y_train})
self.test_loss, self.test_accuracy = sess.run([CNN.loss, CNN.accuracy],
feed_dict={CNN.X: self.X_test,
CNN.Y: self.Y_test})
Train the model

We need now to train the model, evaluate it and test its performance on test
data, and eventually include hyperparameters.
epochs = 100
batch_size = 100
n_filters = 10
n_neurons_connected = 50
n_categories = 10
59
CNN_tf = np.zeros((len(eta_vals), len(lmbd_vals)), dtype=object)
CNN = ConvolutionalNeuralNetworkTensorflow(X_train, Y_train, X_test, Y_test,
n_filters=n_filters, n_neurons_connected=n_neurons_connected
n_categories=n_categories, epochs=epochs, batch_size=batch_s
eta=eta, lmbd=lmbd)
CNN.fit()
print("Test accuracy: %.3f" % CNN.test_accuracy)
print()
CNN_tf[i][j] = CNN
Visualizing the results

sns.set()

CNN = CNN_tf[i][j]
train_accuracy[i][j] = CNN.train_accuracy
test_accuracy[i][j] = CNN.test_accuracy

plt.show()

plt.show()
Running with Keras

60
from tensorflow.keras.layers import Conv2D
from tensorflow.keras.layers import MaxPooling2D
from tensorflow.keras.layers import Flatten
def create_convolutional_neural_network_keras(input_shape, receptive_field,

n_filters, n_neurons_connected, n_categories,
eta, lmbd):
model = Sequential()
model.add(Conv2D(n_filters, (receptive_field, receptive_field), input_shape=input_shape, paddi
activation=’relu’, kernel_regularizer=l2(lmbd)))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Flatten())
model.add(Dense(n_neurons_connected, activation=’relu’, kernel_regularizer=l2(lmbd)))
model.add(Dense(n_categories, activation=’softmax’, kernel_regularizer=l2(lmbd)))
sgd = SGD(lr=eta)
model.compile(loss=’categorical_crossentropy’, optimizer=sgd, metrics=[’accuracy’])
return model
epochs = 100
batch_size = 100
input_shape = X_train.shape[1:4]
receptive_field = 3
n_filters = 10
n_neurons_connected = 50
n_categories = 10
Final part
CNN_keras = np.zeros((len(eta_vals), len(lmbd_vals)), dtype=object)

CNN = create_convolutional_neural_network_keras(input_shape, receptive_field,
n_filters, n_neurons_connected, n_categories,
eta, lmbd)
CNN.fit(X_train, Y_train, epochs=epochs, batch_size=batch_size, verbose=0)
scores = CNN.evaluate(X_test, Y_test)
CNN_keras[i][j] = CNN

print("Test accuracy: %.3f" % scores[1])
print()
Final visualization
sns.set()
61
CNN = CNN_keras[i][j]
train_accuracy[i][j] = CNN.evaluate(X_train, Y_train)[1]

test_accuracy[i][j] = CNN.evaluate(X_test, Y_test)[1]

plt.show()

plt.show()
Fun links
1. Self-Driving cars using a convolutional neural network
2. Abstract art using convolutional neural networks
62

Day 4

Uploaded by

Copyright:

Available Formats

Day 4

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Day 4

Uploaded by

Copyright:

Available Formats

Data Analysis and Machine Learning:

Neural networks, from the simple

Jun 26, 2020

c 1999-2020, Morten Hjorth-Jensen. Released under CC Attribution-NonCommercial 4.0

Neural network types

Convolutional Neural Network

Recurrent neural networks

Other types of networks

Why multilayer perceptrons?

which illustrates a basic property of MLPs: The only independent variables

f (x) = c1 f (c2 x + c3 ) + c4 (10)

Matrix-vector notation. We can introduce a more convenient notation for

Matrix-vector notation and activation. The activation of node i in layer

Activation functions. A property that characterizes a neural network, other

Activation functions, Logistic and Hyperbolic ones. The second require-

"""Plots a graph of the squashing function used by a rectified linear

From one to many layers, the universal approximation the-

Deriving the back propagation code for a multilayer per-

Derivatives and the chain rule

Derivative of the cost function

The derivative of this function with respect to the weights is

Bringing it together, first back propagation equation

This is an important expression. The second term on the right handside

Derivatives in terms of zjL

The starting equations.

An interesting consequence of the above equations is that when the activation

Final back propagating equation

with Ml being the number of nodes in layer l, we obtain

This is our final equation.

Setting up the Back propagation algorithm

Thereafter we compute the ouput error δ̂ L by computing all

Then we compute the back propagate error for each l = L − 1, L − 2, . . . , 2 as

Defining the cost function

Again we take the negative log-likelihood to define our cost function:

C(θ̂) = − log P (D | θ̂).

Example: binary classification problem

where we had defined the logistic (sigmoid) function

The Softmax function

∂f (zil ) ∂f (zil ) ∂zjl ∂f (zil ) l−1

For the Softmax function we have

Its derivative with respect to zjl gives

which in case of the simply binary model reduces to having i = j.

Developing a code for doing neural networks with back

1. Collect and pre-process data

2. Define model and architecture

5. Evaluate model performance on test data

Y = (23.7, 22.2, 27.1, 17.5, 21.1)

# ensure the same random numbers appear every time

# display images in notebook

# download MNIST dataset

# flatten the image

# choose some random images to display

for i, image in enumerate(digits.images[random_indices]):

Train and test datasets

# one-liner from scikit-learn library

#X_train, X_test, Y_train, Y_test = train_test_split_numpy(inputs, labels, train_size, test_size)

print("Number of training images: " + str(len(X_train)))

Define model and architecture

If we were building a binary classifier, it would be sufficient with a single neuron