This presentation Neural Network will help you understand what is a neural network, how a neural network works, what can the neural network do, types of neural network and a use case implementation on how to classify between photos of dogs and cats. Deep Learning uses advanced computing power and special types of neural networks and applies them to large amounts of data to learn, understand, and identify complicated patterns. Automatic language translation and medical diagnoses are examples of deep learning. Most deep learning methods involve artificial neural networks, modeling how our brains work. Neural networks are built on Machine Learning algorithms to create an advanced computation model that works much like the human brain. This neural network tutorial is designed for beginners to provide them the basics of deep learning. Now, let us deep dive into these slides to understand how a neural network actually work. Below topics are explained in this neural network presentation: 1. What is Neural Network? 2. What can Neural Network do? 3. How does Neural Network work? 4. Types of Neural Network 5. Use case - To classify between the photos of dogs and cats Simplilearn’s Deep Learning course will transform you into an expert in deep learning techniques using TensorFlow, the open-source software library designed to conduct machine learning & deep neural network research. With our deep learning course, you'll master deep learning and TensorFlow concepts, learn to implement algorithms, build artificial neural networks and traverse layers of data abstraction to understand the power of data and prepare you for your new role as deep learning scientist. Why Deep Learning? It is one of the most popular software platforms used for deep learning and contains powerful tools to help you build and implement artificial neural networks. Advancements in deep learning are being seen in smartphone applications, creating efficiencies in the power grid, driving advancements in healthcare, improving agricultural yields, and helping us find solutions to climate change. With this Tensorflow course, you’ll build expertise in deep learning models, learn to operate TensorFlow to manage neural networks and interpret the results. You can gain in-depth knowledge of Deep Learning by taking our Deep Learning certification training course. With Simplilearn’s Deep Learning course, you will prepare for a career as a Deep Learning engineer as you master concepts and techniques including supervised and unsupervised learning, mathematical and heuristic aspects, and hands-on modeling to develop algorithms. Learn more at: https://www.simplilearn.com

1. What’s in it for you?
APPLICATIONS OF NEURAL NETWORK
TYPES OF NEURAL NETWORK
USE CASE -
HOW DOES NEURAL NETWORK WORK?
What is NEURAL NETWORK?

2. What’s in it for you?
TYPES OF NEURAL NETWORK
USE CASE - to classify between the photos of dogs and cats
HOW DOES NEURAL NETWORK WORK?
What is NEURAL NETWORK?
What can Neural network do?

3. Brief History of Artificial Intelligence
The word ‘Artificial
Intelligence’ coined by John
McCarthy
‘Shakey’ was the first general
purpose mobile robot built
Supercomputer ‘Deep blue’ was
designed which defeated the
world Chess champion in a game
First commercially successful
robotic vacuum cleaner
created
Speech recognition, RPA,
dancing robots, smart homes
and many more to come from
AI
1956 1969 1997 2002 2005-2018
What is a Neural Network?

4. What is Neural Network?
Hi Guys! I heard you want to
know what Neural network is!

5. What is Neural Network?
As a matter of fact, you have
been using Neural network on
a daily basis!

6. What is Neural Network?
When you ask your mobile
assistant to perform a search
for you…..

7. What is Neural Network?
Self driving cars use it…

8. What is Neural Network?
Computer games use it too!

9. What is Neural Network?
Also used in processing the
map images on your phone..

10. What is Neural Network?
Neural network is a system or
hardware that is designed to operate
like a human brain

11. Brief History of Artificial Intelligence
The word ‘Artificial
Intelligence’ coined by John
McCarthy
‘Shakey’ was the first general
purpose mobile robot built
Supercomputer ‘Deep blue’ was
designed which defeated the
world Chess champion in a game
First commercially successful
robotic vacuum cleaner
created
Speech recognition, RPA,
dancing robots, smart homes
and many more to come from
AI
1956 1969 1997 2002 2005-2018
What can Neural
Network do?

12. What can Neural Network do?
Let’s list out the things Neural
network can do for you..

13. What can Neural Network do?
Translate text

14. What can Neural Network do?
Identify faces
Translate text

15. What can Neural Network do?
Identify faces
Translate text
Recognize speech

16. What can Neural Network do?
Identify faces
Translate text
Read handwritten text
Recognize speech

17. What can Neural Network do?
Identify faces
Recognize speech
Translate text
Read handwritten text
Control robots

18. What can Neural Network do?
And a lot of other things…

19. Brief History of Artificial Intelligence
The word ‘Artificial
Intelligence’ coined by John
McCarthy
‘Shakey’ was the first general
purpose mobile robot built
Supercomputer ‘Deep blue’ was
designed which defeated the
world Chess champion in a game
First commercially successful
robotic vacuum cleaner
created
Speech recognition, RPA,
dancing robots, smart homes
and many more to come from
AI
1956 1969 1997 2002 2005-2018
How does a Neural Network
work?

20. How does a Neural Network work?
Now we have come far enough to
understand how Neural network
works

21. How does a Neural Network work?
These are the different layers of a neural network

22. How does a Neural Network work?
Input layer
Picks up input signals and passes
them to the next LAYER

23. How does a Neural Network work?
Hidden layer
Does all kinds of calculations and
feature extraction

24. How does a Neural Network work?
Output layer
This layer delivers the final result

25. How does a Neural Network work?
Now let’s make use of this neural
network and see how it works!

26. How does a Neural Network work?
Wonder how traffic cameras identify vehicles
Registration plate on the road to detect
SpeedingVehicles and those breaking the
LAW...!!

27. How does a Neural Network work?
28*28 pixels of the image is FEd as
input to identify the Registration
plate
Let’s consider the image of this
vehicle and find out what’s on
the number plate
28
28

28. How does a Neural Network work?
Each Neuron has a number called
activation that represents the grayscale
value of the corresponding pixel ranging
from 0 to 1. 1 for white pixel and 0 for
black pixel
28
28
0.82

29. How does a Neural Network work?
Each neuron is lit up when its activation
is close to 1
28
28
0.82

30. How does a Neural Network work?
Pixels in the form of arrays are fed to
the input layer..
28
28
Pixels of car image fed as
input

31. How does a Neural Network work?
Lets name the inputs x1, x2 and x3
respectively..
x1
x2
x3
28
28
Pixels of car image fed as
input

32. How does a Neural Network work?
The input layer passes it to the
hidden layer..
x1
x2
x3
28
28

33. How does a Neural Network work?
x1
x2
x3
28
28
The interconnections are assigned
weights at random..
W11
W12
W13
W14
W15
W16
W17
W18

34. How does a Neural Network work?
x1
x2
x3
28
28
The weights are multiplied with the
input signals and a bias is added to
all of them..
i=1
n
w x + b
i i*
W11
W12
W13
W14
W15
W16
W17
W18

35. How does a Neural Network work?
x1
x2
x3
28
28
i=1
n
w x + b
i i*
The weighted sum of inputs is fed as
input to the activation function to
decide which nodes to fire for
feature extraction..
W11
W12
W13
W14
W15
W16
W17
W18

36. How does a Neural Network work?
x1
x2
x3
28
28
i=1
n
w x + b
i i* As the signal flows within the hidden
layers, the weighted sum of inputs is
calculated and is fed to the
activation function in each layer to
decide which nodes to fireW11
W12
W13
W14
W15
W16
W17
W18
W21
W22
W23
W24
W25
W26
W27
W28
Feature extraction
of the number plate

37. How does a Neural Network work?
Lets look at the different
activation functions before we
move ahead

38. How does a Neural Network work?
0
1
Y
(X)=
1
1 + e-x
SIGMOID FUNCTION
USED WHEN THE MODEL IS PREDICTING PROBABILITY
i=1
n
w x + b
i i*

39. How does a Neural Network work?
0
1
Y
(X)=
1
1 + e-x
SIGMOID FUNCTION
USED WHEN THE MODEL IS PREDICTING PROBABILITY
i=1
n
w x + b
i i*
i=
1
n
w x + b
i i
0
1
Y
(X)=
1, if x>=0
0, if x<0( )
USED WHEN THE OUTPUT DEPENDS ON A THRESHOLD
VALUE
*
THRESHOLD
FUNCTION

40. How does a Neural Network work?
0
1
Y
(X)=
1
1 + e-x
SIGMOID FUNCTION
USED WHEN THE MODEL IS PREDICTING PROBABILITY
i=1
n
w x + b
i i*
i=
1
n
w x + b
i i
0
1
Y
(X)=
1, if x>=0
0, if x<0( )
USED WHEN THE OUTPUT DEPENDS ON A THRESHOLD
VALUE
*
THRESHOLD
FUNCTION
i=
1
n
w x + b
i i*
0
1
Y
(X) = max(X,0)
relu FUNCTION
It gives an output X if X is positive,
0 otherwise

41. How does a Neural Network work?
0
1
Y
(X)=
1
1 + e-x
SIGMOID FUNCTION
USED WHEN THE MODEL IS PREDICTING PROBABILITY
i=1
n
w x + b
i i*
i=
1
n
w x + b
i i
0
1
Y
(X)=
1, if x>=0
0, if x<0( )
USED WHEN THE OUTPUT DEPENDS ON A THRESHOLD
VALUE
*
THRESHOLD
FUNCTION
i=
1
n
w x + b
i i*
0
1
Y
(X) = max(X,0)
relu FUNCTION
It gives an output X if X is positive,
0 otherwise
Hyperbolic
tangent
function
0
1
Y
(X)=
1 - e
1 + e-2x
i=1
n
w x + b
i i*
-1
-2x
Similar to sigmoid function with a range of (-
1,1)

42. How does a Neural Network work?
Now that you know what is an
activation function, lets get back
to the neural network..

43. How does a Neural Network work?
x1
x2
x3
28
28
Finally, the model would predict the
outcome by applying suitable
activation function to the output
layer
Identifies the number on
the plate
W11
W12
W13
W14
W15
W16
W17
W18
W21
W22
W23
W24
W25
W26
W27
W28
W31
W32
W33
W34
08-D-03858
Optical Character Recognition (OCR) is used on
the images to convert it into text in order to
identify what’s written on the plate

44. How does a Neural Network work?
x1
x2
x3
28
28
W11
W12
W13
W14
W15
W16
W17
W18
W21
W22
W23
W24
W25
W26
W27
W28
W31
W32
W33
W34
Error in the output is
backpropagated through the
network and weights are adjusted to
minimize the error rate.This is
calculated by a cost function
Identifies the number on
the plate
Backpropagation

45. How does a Neural Network work?
The output is compared with the
original result and multiple iterations
are done to get the maximum accuracy

46. How does a Neural Network work?
With every iteration, the weights at every
interconnections are adjusted based on
the error…

47. Brief History of Artificial Intelligence
The word ‘Artificial
Intelligence’ coined by John
McCarthy
‘Shakey’ was the first general
purpose mobile robot built
Supercomputer ‘Deep blue’ was
designed which defeated the
world Chess champion in a game
First commercially successful
robotic vacuum cleaner
created
Speech recognition, RPA,
dancing robots, smart homes
and many more to come from
AI
1956 1969 1997 2002 2005-2018
Types of Artificial Neural
Network

48. Types of Artificial Neural Network
Lets see the different types of
neural network

49. Types of Artificial Neural Network
Feedforward neural
network
Simplest form of ANN. Data
travels only in one direction(input
- output)

50. Types of Artificial Neural Network
Feedforward neural
network
Simplest form of ANN. Data
travels only in one direction(input
- output)
Applications
Vision and speech
recognition

51. Types of Artificial Neural Network
Radial basis
function neural
network
This model classifies the data
point based on its distance from a
center point

52. Types of Artificial Neural Network
Radial basis
function neural
network
This model classifies the data
point based on its distance from a
center point
Applications
Power restoration
systems

53. Types of Artificial Neural Network
Kohonen self
organizing neural
network
Vectors of random dimensions are
input to discrete map comprised
of neurons

54. Types of Artificial Neural Network
Kohonen self
organizing neural
network
Vectors of random dimensions are
input to discrete map comprised
of neurons
Used to recognize patterns in
data like in medical analysis
Applications

55. Types of Artificial Neural Network
Recurrent
neural network
The hidden layer saves its output
to be used for future prediction

56. Types of Artificial Neural Network
Recurrent
neural network
The hidden layer saves its output
to be used for future prediction
Text to speech conversion model
Applications

57. Types of Artificial Neural Network
convolution
neural network
The input features are taken in
batches like a filter.This allows
the network to remember an
image in parts!

58. Types of Artificial Neural Network
convolution
neural network
The input features are taken in
batches like a filter.This allows
the network to remember an
image in parts!
Used in signal and image
processing
Applications

59. Types of Artificial Neural Network
modular neural
networkIt has a collection of different neural
networks working together to get
the output

60. Types of Artificial Neural Network
modular neural
networkIt has a collection of different neural
networks working together to get
the output
Still undergoing research
Applications

61. Types of Artificial Neural Network
Feedforward neural
network
Radial basis
function neural
network
Kohonen self
organizing neural
network
Recurrent
neural network
convolution
neural network
modular neural
network

62. Use Case – Problem Statement
Let’s use the system to identify
between a cat and a dog!

63. Use Case – Problem Statement
Classify photos of cats and
dogs using Neural Network
Problem statement

64. Use Case – Problem Statement
The use case will be
implemented on python

65. Use Case – Implementation
#Import the required packages
from keras.models import Sequential
from keras.layers import Conv2D
from keras.layers import MaxPooling2D
from keras.layers import Flatten
from keras.layers import Dense

66. Use Case – Implementation
# Initialising the CNN
classifier = Sequential()
# Convolution
classifier.add(Conv2D(32, (3, 3), input_shape = (64, 64, 3), activation = 'relu’))
# Pooling
classifier.add(MaxPooling2D(pool_size = (2, 2)))
# Adding a second convolutional layer
classifier.add(Conv2D(32, (3, 3), activation = 'relu'))
classifier.add(MaxPooling2D(pool_size = (2, 2)))

67. Use Case – Implementation
# Flattening
classifier.add(Flatten())
# Full connection
classifier.add(Dense(units = 128, activation = 'relu'))
classifier.add(Dense(units = 1, activation = 'sigmoid’))
# Compiling the CNN
classifier.compile(optimizer = 'adam', loss = 'binary_crossentropy', metrics = ['accuracy'])

68. Use Case – Implementation
# Fitting the CNN to the images
test_datagen = ImageDataGenerator(rescale = 1./255)
training_set = train_datagen.flow_from_directory('dataset/training_set',
target_size = (64, 64),
batch_size = 32,
class_mode = 'binary')

69. Use Case – Implementation
# Fitting the CNN to the images
test_set = test_datagen.flow_from_directory('dataset/test_set',
target_size = (64, 64),
batch_size = 32,
class_mode = 'binary')

70. Use Case – Implementation
# Training the model
classifier.fit_generator(training_set,
steps_per_epoch = 8000,
epochs = 1,
validation_data = test_set,
validation_steps = 2000)

71. Use Case – Implementation
# Part 3 - Making new predictions
import numpy as np
from keras.preprocessing import image
test_image = image.load_img('dataset/single_prediction/cat_or_dog_1.jpg',
target_size = (64, 64))
test_image = image.img_to_array(test_image)
test_image = np.expand_dims(test_image, axis = 0)
result = classifier.predict(test_image)
training_set.class_indices
if result[0][0] == 1:
prediction = 'dog'
else:
prediction = 'cat’

72. Use Case – Implementation
# Part 3 - Making new predictions
prediction

73. Use Case – Implementation
So, we have successfully built a neural
network that can distinguish between
photos of a cat and a dog..

74. Use Case – Implementation
The accuracy of the model is around
80% which is quite good…

75. Key Takeaways

76. So what’s
your next step?