Artificial Intelligence

COMPUTER SCIENCE 9618 PAPER 3

 Artificial Intelligence
Artificial intelligence is the ability of a computer or a robot controlled
by a computer to do tasks that are usually done by humans because
they require human intelligence

Dijkstra’s Algorithm
Note: Maps, telephone network, social networking are all represented
by graphs (nodes) and to find the shortest path between nodes we
use Dijkstra’s Algorithm.

distance u (initial) + cost (u till v) < distance (final) (shortest) (stored)

stored shortest distance ← distance u + cost (u till v)

u v
20 10

Path 1
40
0 + 40 < (infinity)
then shortest distance = 40

Path 2
0 + 20 + 10 < 40
then shortest distance = 30
 3 5 2 9 3 8

Initialize base to 0
Initialize nodes to infinity
Make a table show all the steps

T1 T2 T3 T4 T5 T6

4 5 2

3 5 2

3 5 2 11 8 13

3 5 2 9 3 8

Note : There is no fixed answer to this question. It just that you have
to show that you are aware of Dijkstra’s Algorithm
 A* Algorithm
A* algorithm is based on dijkstra, but adds an extra heuristic value
(intelligent guess) on how far we have to go to reach the destination
most efficiently.
HA

HB B 5 7 12
HC
C 4 9 13
HAB
B 4 7 11
HAE
E 7 3 10
HAEF
F 8 3 11
HAES
School 12 0 12
HAEFS School 11 0 11

Home -> A -> E -> F -> School

Machine Learning And Deep Learning

Neural Network

Learn From
Past
Brain Experience

Neurons

Artificial Neural Network

Machines Are Created To Learn From Past Experience (Data)
 Machine Learning
Is a subset of AI in which algorithms are trained and learn from their
past experiences

Tiktok Algorithm

Labeled Data And Unlabeled Data

Unlabeled data: The data to which tag or label is not attached

Labeled data: The unlabeled becomes labeled data the moment a

meaning is attached (label, tag)

Types Of Machine Learning

Supervised Learning
Supervised learning allows data to be collected, or a data output
produced, from the previous experience.
Supervised learning uses sample data with known outputs (in
training), uses labelled input data.
Able to predict future outcomes based on past data.

Unsupervised Learning
Unsupervised machine learning helps all kinds of unknown
patterns in data to be found.
Unsupervised learning only requires input data to be given.
Uses any data, uses unlabeled input data.

Reinforcement Learning
Reinforcement learning is a machine learning technique based on
feedback
In which an agent learns to behave in an environment by
performing the actions and seeing the results of the actions.
For each good action, the agent gets positive feedback (reward)
and each bad action receives negative feedback (punishment).
The agent learns automatically using feedback without any
labelled data.
Using feedback to improve its performance at accomplishing
similar tasks.

Deep Learning
Is a subset of Machine Learning

Deep learning is a type of machine learning inspired by the

structure of the human brain. This structure is called Artificial
Neural Networks.
Extract patterns from data using neural networks.
Machine Learning Analogy: Sorting Fruits Manually
Imagine you are sorting apples and oranges. In Machine Learning,
you set rules like:
If the fruit is round and red → It's an apple.
If the fruit is round and orange → It's an orange.
Here, you manually define the features (color, shape) and the
system classifies based on those rules. It works well but depends
on the quality of the rules you set.

Deep Learning Analogy: A Child Learning to Recognize Fruits

Now, imagine a child who sees many apples and oranges but isn’t
given any rules. Instead, they learn by observing patterns:
They see different shapes, colors, and textures.
Over time, they naturally figure out which is an apple and which is
an orange without explicit rules.
This is Deep Learning, the system learns patterns by itself, just
like the child, without needing manual feature selection.
Artificial neural networks are based on the interconnection between
neurons in the human brain. The system is able to think like a human
using these neural networks, and its performance improves with
more data.

The hidden layer is where data from the input layer is processed into
something which can be sent to the output layer.

State the reason for having multiple hidden layers in an artificial

neural network?

Enables deep learning to take place.

When the problems you are trying to solve have a higher level of
complexity, it requires more layers to solve.
To enable the neural network to learn and make decisions on its
own.
To improve the accuracy of results.
Back Propagation And Regression

Neural Networks Take Random Weight Values

 Training program is iterative
The errors are propagated back into the neural networks
in order to update the initial networking weightings
This training process is repeated until the desired outputs are
eventually
obtained

Describe the Backpropagation process?

The initial outputs from the system are compared to the expected
output, and the error is calculated.
The system weightings are adjusted to minimize the difference
between actual and expected results.
This process is repeated multiple times, gradually reducing errors.
Once the errors are minimized sufficiently, the neural network is
well-trained and produces accurate outputs.

Describe regression?

Is one of the ways of analyzing data before it is input into a

system.
Is used to make predictions from given data by learning some
relationship between the input and the output.
Explain how artificial neural networks enable machine learning ?

Artificial neural networks are intended to replicate the way

human brains work.
Weights are assigned for each connection between nodes.
The data are input at the input layer and are passed into the
system.
They are analyzed at each subsequent hidden layer, where
outputs are calculated.
This process of learning is repeated many times to achieve
optimum outputs.
Decisions can be made without being specifically programmed.
The deep learning network will have created complex feature
detectors.
The output layer provides the results.
Backpropagation of errors will be used to correct any errors that
have been made

What is meant by Deep Learning?

Uses artificial neural network(s)

that contain(s) a high number of hidden layers
modelled on the human brain
Deep learning uses many layers to progressively extract higher-
level features from the (raw) input
Deep learning is a specialised form of machine learning

Outline the reasons for using Deep Learning?

Deep learning makes good use of unstructured data.

Deep learning outperforms other methods if the data size is large.
Deep learning systems enable machines to process data with a
nonlinear approach.
Deep learning is effective at identifying (hidden) patterns /
patterns that humans might not be able to see / patterns that are
too complex or time-consuming for humans to carry out.
It can provide a more accurate outcome with higher numbers of
hidden layers.
Describe the purpose of both A* and Dijkstra algorithm

To find the optimal, shortest or most cost-effective route

between two nodes
based on distance / cost / time.

Explain what is meant by the term artificial network ?

An artificial neural network is a component of artificial

intelligence that is meant to simulate the functioning of a
biological brain.
Artificial neural networks are a key component of machine
learning.
They can solve problems that would prove impossible or difficult
for humans.
Artificial neural networks have self-learning capabilities that
enable them to produce better results as more data becomes
available.
Artificial neural networks can be layered (input, hidden, and
output layers).
Artificial neural networks have many interconnected layers,
some/many of which are hidden.
Weights are assigned between nodes.
Weights are adjusted through training to give a more accurate
result.
More complex learning capabilities / more accurate results are
available with larger numbers of hidden layers.
Explain the use of graphs to Aid Artificial Intelligence ?

Artificial Neural Networks can be represented using graphs

Graphs provide relationships between nodes
AI problems can be defined/solved as finding a path in a graph
Graphs may be analyzed by a range of algorithms
e.g. A* / Dijkstra’s algorithm
used in machine learning.
Example of method e.g. Back propagation of errors / regression
methods
 Artificial Intelligence
Questions

Question 1 (9618_s21_qp_31)

Sir Taha Ali Papersdock +92 318 2248934

Sir Taha Ali Papersdock +92 318 2248934
Question 2 (9618_w21_qp_31)

Sir Taha Ali Papersdock +92 318 2248934

Sir Taha Ali Papersdock +92 318 2248934
Question 3 (9618_w22_qp_31)

Question 4 (9618_w22_qp_32)

Sir Taha Ali Papersdock +92 318 2248934

Question 5 (9618_s23_qp_31)

Sir Taha Ali Papersdock +92 318 2248934

Question 6 (9618_s23_qp_32)

Sir Taha Ali Papersdock +92 318 2248934

Question 7 (9618_w23_ms_31)

Question 8 (9618_w23_qp_32)

Sir Taha Ali Papersdock +92 318 2248934

Question 9 (9618_s24_qp_31)

Question 10 (9618_s24_qp_32)

Sir Taha Ali Papersdock +92 318 2248934

 Answers
Answer 1 (9618_s21_ms_31)

Sir Taha Ali Papersdock +92 318 2248934

Answer 2 (9618_w21_ms_31)

Sir Taha Ali Papersdock +92 318 2248934

Sir Taha Ali Papersdock +92 318 2248934
Answer 3 (9618_w22_ms_31)

Answer 4 (9618_w22_ms_32)

Sir Taha Ali Papersdock +92 318 2248934

Answer 5 (9618_s23_ms_31)

Answer 6 (9618_s23_ms_32)

Sir Taha Ali Papersdock +92 318 2248934

Sir Taha Ali Papersdock +92 318 2248934
Answer 7 (9618_w23_ms_31)

Answer 8 (9618_w23_ms_32)

Answer 9 (9618_s24_ms_31)

Sir Taha Ali Papersdock +92 318 2248934

Answer 10 (9618_s24_ms_32)

Sir Taha Ali Papersdock +92 318 2248934