Lecture 0

INT354: MACHINE LEARNING-I

 Course Overview
• L T P: 2 0 2

• Credit: 3

• Text Book:
1. MACHINE LEARNING : A PRACTITIONER'S APPROACH by CHANDRA
S.S., VINOD; HAREENDRAN S., ANAND, PHI Learning
References Book:
1. UNDERSTANDING-MACHINE-LEARNING-THEORY-ALGORITHMS
FROM THEORY TO ALGORITHM by SHAI SHALEV-SHAWARTZ AND
SHAI BEN-DAVID, CAMBRIDGE UNIVERSITY PRESS
2. MACHINE_LEARNING_IN_ACTION by PETER HARRINGTON, Manning
Publications
Course Assessment Model

⚫ Marks break up*

Attendance 5 marks
CA (Test, Test-code based, Project: Mandatory) 25 marks

⚫ MTT(MCQ) 20 marks

⚫ ETT(MCQ) 50 marks

⚫ Total 100 marks

 OER

OPEN EDUCATIONAL RESOURCE

Course Code: INT354
Course Tittle: :MACHINE LEARNING-I
L.T.P: 3.0.0 Credit: 3
Course Code Course Title Unit mappe d Broad topic OER Title of OER *%age unit mapped Source URL
Type with OER
(approx)

Unit 1 Machine Learning, Need Reading material INT354 90%

of Machine Learning, (Pdf ) machine learningand
Types of Learning, typesof machine
Well Posed Learning Youtube link learning
Problems, Designing
a Learning Needof Machine
Systems, Statistical Learning, WellPosed
Learning Framework, Learning Problems,
, Empirical Designinga Learning
Risk Minimization, Systems, Statistical
Empirical Risk Learning Framework, ,
Minimization with Empirical Risk
Inductive Bias, Minimization, Empirical
PAC Learning, Risk Minimization with
Building good training Inductive Bias,PAC
sets Learning, Building good
training sets
INT-354

:MACHINE LEARNING-I

Unit 2 Machine learning Reading material INT 354 70% Machine learning
classifiers : (Pdf ) classifiers :

Unit 3 Generative models: Reading material INT 354 80% Generative models
Maximum (Pdf ) :Maxi mum
Likelihood Likelihood
Estimator,Baye Estimator,
sian Learning,
Bayes Theorem, BruteForce Bayesian Learning, Bayes
Concept Learning, Bayes Theorem, BruteForce
Optimal Classifier,, Gibs Concept Learning, Bayes
Algorithm, Naive Optimal Classifier,
Bayes Classifier, EM

Model evaluation
, Gibs
and hyperparameter tuning,
Algorithm, NaiveBayes
:
Classifier,
Streamlining Workflows with
EM,
Pipelines,

Model evaluation and

Using kfold Cross
hyperparamet er tuning,
Validation to Access
Model Performance,

: Streamlining Workflows
with Pipelines,
Debugging Algorithms with
Learning and Validation Curves

Usingkfold Cross Validation

to Access Model
Fine-Tuning Machine Learning
Performance,
Models via Grid Search

Debugging
 Algorithms
with Learning
and Validation Curves,

Fine-Tuning Machine
Learning Modelsvia Grid
Search

Unit 4 Predicting Reading INT 354 80% Predicting

continuous target variables with material (Pdf ) continuous target variables
regression analysis with regression analysis:
: Introducing Linear Introducing Linear
Regression, Fitting a Robust Regression, Fittinga Robust
Regression Model Regression Modelusing
using RANSAC, RANSAC,

Relationship Using a Relationship Usinga

Correlation Matrix, Correlation Matrix,

Regularized Methods
for
 Regression, Regularized
Polynomial Methodsfor
Regression, Decision Tree, Regression, Polynomial
ARIMA Regression, Decision Tree,
ARIMA
Unit 5 Regression Reading INT 354 90% Regression
Metrics : R2 material (Pdf ) Metrics : R2
Score, Mean Absolute Error, Score, Mean Absolute Error,M
Mean Squared Error, ean Squared Error,Mean Squar
Mean Squared ed Logarithmic Error,Mean
Logarithmic Error, Absolute Percentage Error,
Mean Absolute Explained Variance Score,D2
Percentage Error, Explained ScoreVisual Evaluation of
Variance Score, D2 Regression Models
Score Visual Evaluation
of Regression Models

Unit 6 The bias- Reading INT 354 80% Thebias-

complexity tradeoff: material (Pdf ) complexity tradeoff,

No Free Lunch Theorem, Error NoFree Lunch Theorem, Error

Decomposition

TheVC- Dimension, The

Rademacher

The VC- Dimension, The

Rademacher
 Complexity, Complexity,
The Natarajan Dimension The Natarajan Dimension

Algorithm- Independent machine Algorithm- Independent

Learning : Combining machine Learning: Combining
Classifiers, Majority Voting Classifiers, Majority Voting
Classifier, Resampling for Classifier, Resampling for
Estimating Statistics, Lack of Estimating Statistics, Lackof
Inherent Superiority Inherent Superiority of
of Classifier, Bagging Classifier, Baggingand
and Boosting Classifier, Random Boosting Classifier, Random
Forest Classifier, Regressor, Forest Classifier, Regressor,
Support Vector Classifier Support Vector Classifier and
and Regressor Regressor

**Average % age of -- --- --- CSE254 Avg.=

total syllabus mapped 81.66%
 Program Outcomes

PO-1 Engineering knowledge::Apply the knowledge of mathematics, science, engineering

fundamentals, and an engineering specialization to the solution of complex engineering problems.
PO-2 Problem analysis::Identify, formulate, research literature, and analyze complex engineering
problems reaching substantiated conclusions using first principles of mathematics, natural sciences,
and engineering sciences.
Program outcomes

PO-3 Design/development of solutions::Design solutions for complex engineering problems and

design system components or processes that meet the specified needs with appropriate
consideration for the public health and safety, and the cultural, societal, and environmental
considerations.
PO-6 The engineer and society::Apply reasoning informed by the contextual knowledge to assess
societal, health, safety, legal and cultural issues and the consequent responsibilities relevant to
the professional engineering practice.
PO-12 Life-long learning::Recognize the need for, and have the preparation and ability to engage
in independent and life-long learning in the broadest context of technological change.
PO-13 Competitive Skills::Ability to compete in national and international technical events and
11
building the competitive spirit
Revised Bloom’s taxonomy (RBT)

12
 What are Cohorts

⚫ A group of students of a common programme

who intend to attain similar characteristics by
means of learning similar skills in order to target a
particular career opportunity.

13
 Purpose of Cohorts
⚫ Student shall be able to have a goal oriented approach for
his/her career
⚫ Student identifies the goal in the very first year
⚫ Student shall be able to follow the stage wise career
progression.
⚫ Early identification of skill set required for selected goal.

14
INT354 –
Cohort 2 and 5
Course Outcomes
CO1 :: Explore Different types of Machine Learning and statistics used for risk minimization.

CO2 :: Analyze the operations of different types of Machine Learning Classifiers.

CO3 :: Examine the performance of Generative models based on Bayesian learning to solve different
classification problems.

CO4 :: Develop the model that predict value of continuous variable with regression analysis.

CO5 :: Discuss the methods for Error calculations using different Regression metrics.

CO6 :: Extend the Machine Learning approach to understand the bias complexity tradeoff and algorithm
independent machine learning.
What is Machine Learning?
Why is it important??

https://youtu.be/Kjfz8s_d5HM?si=_JfXcKos7P897ROJ
Make Machine your Friend
Overview of Unit 1
 Unit 1
Introduction to machine learning : Machine Learning, Need of Machine Learning,
Types of Learning, Well Posed Learning Problems, Designing a Learning Systems,
Statistical Learning Framework, Empirical Risk Minimization, Empirical Risk
Minimization with Inductive Bias, PAC Learning

Building good training sets : Data Preprocessing, Dealing with Missing Data, Handling
Categorical Data, Partitioning a Dataset in Training and Test Sets, Normalization,
Selecting Meaningful Features
Overview of Unit 2
 Unit 2
Machine learning classifiers : Motivation: When One Variable Is Not
Enough, Choosing a Classification Algorithm, First Steps with Scikit-Learn,
Perceptron Classifier, Stochastic Gradient Descent, Modeling Class
Probabilities via Logistic Regression, Maximum Margin Classification with
Support Vector Machine, Decision Tree Learning, CART, ID3, C4.5, Density
Estimation, Parzen Window, The Nearest Neighbour Rule, KNearest
Neighbour Estimation
Unit 3
Generative models : Maximum Likelihood Estimator,
Bayesian Learning, Bayes Theorem, BruteForce Concept
Learning, Bayes Optimal Classifier, Gibs Algorithm, Naive
Bayes Classifier, EM Algorithm

Model evaluation and hyperparameter tuning :

Streamlining Workflows with Pipelines, Using kfold Cross
Validation to Access Model Performance, Debugging
Algorithms with Learning and Validation Curves, Fine-
Tuning Machine Learning Models via Grid Search
Unit 4
Predicting continuous target variables with regression
analysis : Introducing Linear Regression, Fitting a Robust
Regression Model using RANSAC, Relationship Using a
Correlation Matrix, Exploratory Data Analysis,
Regularized Methods for Regression, Polynomial
Regression, Decision Tree, ARIMA
 Unit 5
Regression Metrics : R2 Score, Mean Absolute Error, Mean
Squared Error, Mean Squared Logarithmic Error, Mean Absolute
Percentage Error, Explained Variance Score, D2 Score Visual
Evaluation of Regression Models
Unit 6
The bias-complexity tradeoff : No Free Lunch Theorem,
Error Decomposition, The VC-Dimension, The
Rademacher Complexity, The Natarajan Dimension

Algorithm-Independent machine Learning : Combining

Classifiers, Majority Voting Classifier, Re-sampling for
Estimating Statistics, Lack of Inherent Superiority of
Classifier, Bagging and Boosting Classifier, Random Forest
Classifier, Regressor, Support Vector Classifier and
Regressor