What Is A Neural Network? | How Deep Neural Networks Work | Neural Network Tutorial | Simplilearn

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?

What’s in it for you?
How does neural network work?
Advantages of neural network
Applications of neural network
What is Artificial nEural Network?
Future of Neural Network
What is Deep Learning?

Brief History of Artificial Neural Network
Hello!

I am the human brain

I am the most complex organ in the
human body and I help you to think,
understand and take decisions!

And the secret behind all of my
power is a neuron, I’ll get back
to it in some time!

Ever since the 1950’s, scientists have been
trying to mimic the functioning of a neuron
and use it to build smarter robots!

After a lot of trial and error, humans finally
designed a computer that can recognize
human speech!!

It was only after 2000 that humans were
able to give birth to deep learning that was
able to see and distinguish between
different images and videos!

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

SO what exactly is Deep Learning?

Deep learning is a machine learning technique that teaches computers to do what
comes naturally to humans: learn by example
These are the
images of ‘Dogs’
The robot gets trained with
photos as example!

Wait! I know what you are

You are a dog!
Woof woof!
And is able to identify the
thing that it saw in the
images!

That brings us to the question, how does
deep learning do it?

Remember the neuron?

Scientist managed to build an artificial
form of it that powers any deep learning
based machine!

McCarthy
created
AI
1956 1969 1997 2002 2005-2018
What is Artificial Neural Network?

To understand how an artificial neuron
works, we need to understand how the real
one works!

Dendrite
Axon
Input to a neuron
Cell body
Information processing
happens here
Axon
Output to the
neuron

Now, let’s see what an artificial neural
network looks like

Input layer
Hidden layers
Output layer

So, you can see how similar they are!
Input
Feature
extraction is done
here
Output
Input to a
neuron
Information processing
happens here
AxonOutput to the
neuron

Input layer
Hidden layers
Output layer
a neural network is a system of hardware and/or software patterned after the operation of neurons in the human
brain. Neural networks also called artificial neural networks is a way of achieving deep learning
Artificial neural network
Input to a neuron
Information processing happens here
Axon
Output to the neuron
Biological neural network

McCarthy
created
AI
1956 1969 1997 2002 2005-2018
How does Artificial Neural Network
work?

How does Artificial Neural Network work?
Let us find out how does an artificial neural
network work!

Hey Siri,What is the time
now?

It’s 12:30 in the morning!

Thanks!

Let’s find out how she
recognizes speech!!

Here is a neural network and the
different layers on it
Input layer
Hidden layers
Output layer
a
b
c
D
w
x
y
z
....

This is the sentence that
needs to be recognized by the
network
Input layer
Hidden layers
Output layer
a
b
c
D
w
x
y
z
....

is
What
the
Time?
Input layer
Hidden layers
Output layer
a
b
c
D
w
x
y
z
....

is
What
the
Time?
The sentence is sampled into discrete
sound waves
Input layer
Hidden layers
Output layer
a
b
c
D
w
x
y
z
....

is
What
the
Time?
First, lets consider the word ‘What’
Input layer
Hidden layers
Output layer
a
b
c
D
w
x
y
z
....

What
The waveform is split based on every
letter
W H a t
Input layer
Hidden layers
Output layer
a
b
c
D
w
x
y
z
....

What
W H a t
Now, we will split the soundwave for the
letter ‘w’ into smaller segments..
Input layer
Hidden layers
Output layer
a
b
c
D
w
x
y
z
....

W H a t
As the amplitude is varying in the sound wave for ‘w’,
we collect the values at different intervals and form an
array
Input layer
Hidden layers
Output layer
a
b
c
D
w
x
y
z
....
0.5
1.5
1.7
1.9

W H a t
We feed the array of amplitudes to the
input layer
Input layer
Hidden layers
Output layer
a
b
c
D
w
x
y
z
....
0.5
1.5
1.7
1.9
x1
x2
x3
x4

W H a t
Input layer
Hidden layers
Output layer
a
b
c
D
w
x
y
z
....
W1
W2
W3
W4
W5
W6
W7
W8
W9
w10
Random weights are assigned to each
interconnection between input and hidden
layer
0.5
1.5
1.7
1.9
x1
x2
x3
x4

W H a t
Input layer
Hidden layers
Output layer
a
b
c
D
w
x
y
z
....
W1
W2
W3
W4
W5
W6
W7
W8
W9
w10
i=1
n
w x + b
i i*
The weights are multiplied with the inputs
and a bias is added to form the transfer
function
0.5
1.5
1.7
1.9
x1
x2
x3
x4

W H a t
Input layer
Hidden layers
Output layer
a
b
c
D
w
x
y
z
....
i=1
n
w x + b
i i*
Weights are assigned to the
interconnection between the
hidden layers
W11
W12
W13
W14
W15
W16
W17
W18
W19
W20
W21
W22
0.5
1.5
1.7
1.9
x1
x2
x3
x4

W H a t
Input layer
Hidden layers
Output layer
a
b
c
D
w
x
y
z
....
i=1
n
w x + b
i i*
W11
W12
W13
W14
W15
W16
W17
W18
W19
W20
W21
W22
the output of the transfer function is fed
as an input to the activation function
0.5
1.5
1.7
1.9
x1
x2
x3
x4

Accoustic
model
Acoustic model contains the statistical
representation of each distinct sound that makes a
word
W H a t
Input layer
Hidden layers
Output layer
a
b
c
D
w
x
y
z
....
0.5
1.5
1.7
1.9
x1
x2
x3
x4

Accoustic
model
lexiconLexicon contains the data for different
pronunciations of every word
W H a t
Input layer
Hidden layers
Output layer
a
b
c
D
w
x
y
z
....
0.5
1.5
1.7
1.9
x1
x2
x3
x4

Accoustic
model
lexicon
Language model contains a large list of
words and the probability of them occurring
W H a t
Input layer
Hidden layers
Output layer
a
b
c
D
w
x
y
z
....
Language
model
0.5
1.5
1.7
1.9
x1
x2
x3
x4

With the help of these models, the network
predicts the words that you are trying to say!
Acoustic
model
lexicon
W H a t
Input layer
Hidden layers
Output layer
a
b
c
D
w
x
y
z
....
Language
model
output
0.5
1.5
1.7
1.9
x1
x2
x3
x4

With the help of these models, the network
predicts the words that you are trying to say!
Accoustic
model
lexicon
W H a t
Input layer
Hidden layers
Output layer
a
b
c
D
w
x
y
z
....
Language
model
output
0.5
1.5
1.7
1.9
x1
x2
x3
x4

is
What
is
What
the
Time?
the
Time?
Following the same process for every word and letter,
the neural network recognizes the sentence you said..

Input layer
Hidden layers
Output layer
is
What
the
Time?
x1
x2
x3
x4
W23
W24
W25
W26
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
T
U
V
W
X
Y
Z
is
What
the
Time?
It’s 12:30!

McCarthy
created
AI
1956 1969 1997 2002 2005-2018
Advantages of Artificial Neural
Network

Advantages of Artificial Neural Network
So gentlemen, could you tell me
the advantages of Artificial Neural
Network?

an artificial neural network’s outputs aren't
limited entirely by inputs and results given to
them initially by an expert system

This ability comes in handy for robotics
and pattern recognition systems!

Artificial Neural Networks have the
potential for high fault tolerance!

Artificial Neural Network are capable of
debugging or diagnosing a network on
their own!

Non - linear systems have the capability of
finding shortcuts to reach computationally
expensive solutions

McCarthy
created
AI
1956 1969 1997 2002 2005-2018
Applications of Artificial
Neural Network

Applications of Artificial Neural Network
Let’s look at some of the real life applications
of artificial neural network!

Handwriting recognition
Neural network is used to convert handwritten
characters into digital characters that the system
can recognize

Stock exchange prediction
There are many factors that effect the stock
market. Neural network can examine a lot of
factors and predict the prices on a daily basis
helping the stock brokers

Travelling salesman problem
It refers to finding the optimal path to travel
between all cities in an area. Neural network
helps solve this problem providing higher
revenue at a minimal cost

Image compression
Idea behind data compression neural networks
is to store, encrypt and re-create the actual
image again.

McCarthy
created
AI
1956 1969 1997 2002 2005-2018
Future of Artificial Neural
Network

Future of Artificial Neural Network
I’ll tell you what I see in the
future!

More personalized choices
for users and customers all
over the world

over the world
Hyper intelligent virtual
assistants will make life
easier

over the world
easier
New forms of algorithm
or learning methods
would be discovered

over the world
easier
or learning methods
would be discoveredNeural networks will be
a lot faster in the future

over the world
easier
or learning methods
Neural network tools will
be embedded in every
design surface

over the world
easier
or learning methods
Neural network tools will
be embedded in every
design surface
Neural networks will be used
in the field of medicine,
agriculture, etc

What Is A Neural Network? | How Deep Neural Networks Work | Neural Network Tutorial | Simplilearn

Recommended

More Related Content

What's hot (20)

Similar to What Is A Neural Network? | How Deep Neural Networks Work | Neural Network Tutorial | Simplilearn (20)

More from Simplilearn (20)

Recently uploaded (20)

What Is A Neural Network? | How Deep Neural Networks Work | Neural Network Tutorial | Simplilearn

Editor's Notes