CS001:- Machine Learning ML In Depth + 1 Month

Internship :-

Section - 1 [ Toolkit for Machine Learning ]

❖ Numpy
❖ Pandas
❖ Matplotlib
❖ Mathematics for Machine Learning
❖ Quizzes

Problem Set - 0 Releases

Section - 2 [ Basics of Machine Learning ]

❖ Fundamentals of Machine Learning

➢ What is Machine Learning? The way I like to think about it!
➢ Cool Applications of Machine Learning
➢ Types of ML and Their Types
➢ Workflow of basic ML Problem
➢ Main Challenges of Machine Learning
➢ Dividing the data
➢ Two famous problems of Machine learning:-
■ Underfitting
■ Overfitting
➢ Solutions to the Overfitting and Underfitting
➢ Quizzes

Problem Set Releases

Section - 3 [ Linear Regression & Regularization ]

❖ Linear Regression
➢ What is Linear Regression? Visual Understanding
➢ Hypothesis Function Or Prediction Function
➢ Closed Form Solution aka Normal Equation
➢ Coding Normal Equation
➢ Cost Function
➢ Gradient Descent
 ➢ Assumptions & Pros and Cons of Linear Regression
➢ Proposal for Project
➢ Milestone Project:- Advanced House Price Predictor

❖ Regularized Linear Models

➢ Ridge Regression
➢ Lasso Regression
➢ Elastic Net

● Linear Regression Programming Assignment

● Problem Sets Releases
● Programming Assignment:- Regularized Linear Models

Section - 4 [ Logistic Regression & Performance Metrics ]

❖ Logistic Regression
➢ Hypothesis Function
➢ Cost Function
➢ Gradient Descent
➢ Assumptions and Pros and Cons
➢ Project Proposal
➢ Milestone Project:- Building a Breast Cancer Detection System With
Deployment

❖ Performance Models
➢ MSE
➢ MAE
➢ RMSE
➢ Confusion Matrix ( TP, FP, TN, FN )
➢ Precision and Recall
➢ Understanding the plots
➢ AUC and ROC Curve

● Programming Assignments ( 2 assignments )

● Problem Sets Releases ( 2 sets )

Section - 5 [ Support Vector Machine ]

❖ Support Vector Machines

 ➢ Linear SVM Classification
■ Hard/Soft Margin Classification
➢ Non-Linear SVM Classification
■ How does it work?
■ Polynomial Kernel [Homogenous & Inhomogeneous ]
■ RBF Kernel
➢ Computing SVM Classifier
■ Primal and Dual Problem
■ Sub-Gradient Descent
■ Coordinate Descent
➢ Transductive SVM
➢ SVR

● SVM Programming Assignment

● Problem Sets releases

Section - 6 [ PCA ]
➢ Review of Linear Transformation & EigenVectors and EigenValues
➢ Dimensionality Reduction Need
➢ Basic Intuition Behind PCA
➢ Algorithm:-
■ Data Preprocessing [Data Standardization]
■ Compute the Covariance Matrix
■ Compute the cumulative energy content for each eigenvector
■ Select a subset of the eigenvectors as basis vectors
■ Projecting Back
➢ Pros and Cons
➢ Summary

❖ PCA Programming Assignment

❖ Problem Sets releases

Section - 7 [ Learning Theory ]

❖ Bias and Variance TradeOff

❖ Approx Estimation Error
❖ Empirical Risk Minimization
❖ Empirical Risk Minimization with Inductive Bias

❖ Problem Sets releases

Section - 8 [ Decision Trees & Random Forest ]

❖ Decision Trees
➢ Training of Decision Trees
➢ Prediction in Decision Trees
➢ Information Measures:-
■ Entropy
■ Information Gain
■ Gini Impurity
➢ Hyperparameter Tuning
➢ Project Proposal

● Flowers Species Detection using Trees

● Decision Trees Assignment
● Problem Sets releases

❖ Ensemble Learning
➢ Ensemble Learning
■ Bagging
● Random Forest
■ Boosting
● Gradient Boosting
● Adaboost
● XGboost
■ Stacking
■ Cascading

➢ Ensemble Learning Assignments [ 3 assignments ]

➢ Problem Sets releases [ 3 problems sets ]
➢ Project Proposal
➢ Disease Prediction With deployment
➢ Heart Failure Detection With deployment

Section - 9 [ Naive Bayes ]

❖ Naive Bayes
➢ Bayes Theorem
➢ Simple Naive Bayes
➢ Gaussian Naive Bayes
➢ Bernoulli Naive Bayes
 ➢ Multinomial Naive Bayes
➢ Designing a Document Classification System Mathematically
➢ Naive Bayes Assignment:- Design a Person Classification System
Mathematically
➢ Project Proposal
➢ Spam Detection System Using Naive Bayes
➢ Fake News Detection System

Section - 10 [ Unsupervised Learning Algorithms ]

❖ Unsupervised Learning and Clustering Intro and Types

❖ Cluster Analysis
❖ Unsupervised Learning Algorithms
➢ K-Means with K-Means ++
➢ Hierarchical Clustering Techniques
➢ DBSCAN
➢ OPTICS
➢ Anomaly Detection [ LOCAL OUTLIER FACTOR ]
➢ Unsupervised Learning Problem set and project releases
➢ K-Means Programming Assignment

❖ Customer Segmentation Project

Section - 11 [ Neural Networks ]

➢ Session 1:- Intro to Neural Networks

■ Perceptron and MLP
■ Including both forward and Backward Prop
➢ Session 2:- NLP
■ Building a Sentiment Analysis and Fake New Classifier
➢ Session 3:- Computer Vision
■ Intro about yourself
■ Basics of CNN
■ Advanced CNN
■ Building a Cat and Dog Classifier

Section - 12 [ Writing a Thesis Paper ]

➢ Students have to choose in which topic they will write the paper in latex
and they will follow the full procedure for writing a paper.
Note:-
● In every learning algorithm, assumptions, pros, cons will be there.
● Students will be working as an Intern on various projects for Real world AI
Products with the peers, Certificate of Appreciation will be provided only after
students complete the 1 month internship.

● When Student will enter in the course, he will get private code to access the
account of gradescope to submit the assignments and get feedback.

● Students will get a private group which is like LMS where everything will be
there.

● Students have to submit the assignments before the deadline, otherwise grading
will decrease and if it decreases then you will be not able to do the internship or
you will lose your certificates and LORS. So, Grading is MUST.

● Attending classes is a must, otherwise 2pts will be deducted and if you want leave
then you have to write an application for 1 day leave.

● Students will get a private schedule of the classes and the private zoom pro link.
The lectures are recorded with editing