M AC H I N E L E A R N I N G

INTRODUCTION


Machine Learning
Arthur Samuel, a pioneer in the field of artificial intelligence and computer gaming, coined
the term “Machine Learning” as – “Field of study that gives computers the capability
to learn without being explicitly programmed”.

How it is different from traditional

Programming:
➢ In Traditional Programming, we feed the Input,
Program logic and run the program to get
output.
➢ In Machine Learning, we feed the input, output
and run it on machine during training and the
machine creates its own logic, which is being
evaluated while testing.
 Terminologies that one should know before starting Machine Learning:

❑ Model: A model is a specific representation learned from data by applying some

machine learning algorithm. A model is also called hypothesis.

❑ Feature: A feature is an individual measurable property of our data. A set of numeric

features can be conveniently described by a feature vector. Feature vectors are fed as
input to the model. For example, in order to predict a fruit, there may be features like
color, smell, taste, etc.

❑ Target(Label): A target variable or label is the value to be predicted by our model. For
the fruit example discussed in the features section, the label with each set of input
would be the name of the fruit like apple, orange, banana, etc.
❑ Training: The idea is to give a set of inputs(features) and it’s expected outputs(labels),
so after training, we will have a model (hypothesis) that will then map new data to one
of the categories trained on.

❑ Prediction: Once our model is ready, it can be fed a set of inputs to which it will
provide a predicted output(label).
 M AC H I N E L E A R N I N G

ML


What is learnt How is it learnt How is it

converted to
Learning action
 Decision making
 Hypothesis
Induction, Actions,
generalization guesses

GENERAL INDUCTIVE LEARNING

Feedback,
Observations Refinement
more
observations
S T E P S I N B U I L D I N G A M AC H I N E
LEARNING SYSTEM
• Define Objective
define the purpose or the objective you want to accomplish using your system. This is an important step since the data
you will collect, the algorithm you use, and many other factors depend on this objective.
• Collect Data
you should collect the required data. It is a time-consuming process, but it is the next important step that you must
achieve. You should collect the relevant data and ensure that it is the right data for the problem you are trying to solve.
• Prepare Data
It is only when the input data is clean and relevant that you will obtain an accurate result or prediction.
• Select Algorithm
Numerous algorithms can be used to solve a problem, including Structured Vector Machine (SVM), k-nearest, Naive-
Bayes and Apriori, etc. You must choose the algorithm that best suits the objective.
• Train Model
When your data set is ready, you should feed it into the system and help the machine learn using the chosen algorithm.
• Test Model
When your model is trained, and you believe that it has provided the relevant results, you can test the accuracy of the model
using a test data set.
• Predict
The model will perform numerous iterations with the training data set and the test data set. You can look at the predictions
and provide feedback to the model to help it improve the predictions that it makes.
• Deploy
This means that the model that you have developed can now be deployed.
 H OW D O W E L E A R N ?

Human Machine
Memorize k-Nearest Neighbors,
Case-based learning
Observe someone else, then repeat Supervised Learning, Learning by
Demonstration
Keep trying until it works (riding a bike) Reinforcement Learning

20 Questions Decision Tree

Pattern matching (faces, voices, languages) Pattern Recognition

Guess that current trend will continue Regression

(stock market, real estate prices)
M AC H I N E L E A R N I N G



W H Y U S E M AC H I N E L E A R N I N G ?






M AC H I N E L E A R N I N G I N AC T I O N


A S S O C I AT I O N R U L E S


 C L A S S I F I C AT I O N

 Credit scoring
 Goal: label each
person as
“high risk” or “low risk”
 Input features:
Income and Savings

 Learned discriminant:
If Income > θ1 AND Savings > θ2
THEN low-risk ELSE high-risk
C L A S S I F I C AT I O N M E T H O D S I N T H I S C O U R S E






REGRESSION

 y = wx+w0



• Supervised learning
TYPES OF
• Unsupervised learning M AC H I N E
LEARNING
• Semi-supervised learning

• Reinforcement learning
 • Supervised learning



 • Un-Supervised learning


SEMI-SUPERVISED LEARNING


 • REINFORCEMENT LEARNING



R E L AT I O N S H I P B E T W E E N M A C H I N E
LEARNING AND OTHER COMPUTER
SCIENCE DISCIPLINES
M A C H I N E L E A R N I N G A N D G E N E R AT I V E A I
K E Y A D VA N C E M E N T S
H AV E P R O P E L L E D
THE EVOLUTION OF
G E N E R AT I V E A I

Diffusion Models: These learn to create new data by gradually reversing a process of adding noise to existing
data. They have proven exceptionally powerful in generating high-resolution images and other complex media
 T H E I M P O R TA N C E O F M A C H I N E L E A R N I N G

•
•
•
WHEN DO WE NEED MACHINE LEARNING?

•
•
•
•
•
•
 Soft skills are what set apart effective machine learning

MACHINE LEARNING professionals from those who are ineffective. Such skills include
SKILLS communication, problem-solving, time management, teamwork,
and a thirst for learning.

• Software Engineering Includes the ability to write computer programs, understanding of algorithms and
data structures, and knowledge of computer architecture and organization.

• Statistics and Mathematics This entails having proficiency in hypothesis testing, data modeling, and a strong
grasp of mathematical concepts such as probability, statistics, and linear algebra. It also involves the ability
to devise an evaluation strategy for predictive models and algorithms.
 W H AT D O M A C H I N E L E A R N I N G P R O F E S S I O N A L S D O ?


 R E A L - W O R L D A P P L I C AT I O N S O F M A C H I N E L E A R N I N G


 MACHINE LEARNING AND ETHICAL CONCERNS

Addressing these concerns collectively will guarantee that machine learning evolves and is
deployed to align with societal values and ethical standards, ultimately serving the greater good
T H A N K YO U