Integrated Program in Data Science, Artificial Intelligence & Machine Learning 

Python Programming

Installation and set-up, Python Basics - Variables, built-in functions (print, type, help, range, input,...)

Loops and simple operations, Data Structures and Operations (Lists and Tuple)

Data Structures and Operations (Dictionary, Set), Conditional Statements, Functions and methods

Python Modules overview - Pandas, Numpy - DataFrames & Arrays. Functions, Numpy Array Operations, OOPs and Debugging concepts, Overview of ML Libraries

Data analysis with Python

Pandas Basics - DataFrames & Arrays, reading files, row/column selection, sub-setting, EDA, variable creation, data summaries

Pandas Advanced - Visualization with matplotlib and Seaborn, Data profiling and analysis, variable correlation. Handling data anomalies, feature engineering

ML Mathematics

Mathematics Basics and Advanced

Linear Algebra: Vectors and Scalars, Matrices and Matrix operations, 2D/3D Plots, Functions, Limits and Derivatives, Notations, Numbers, Sequences, Points, Lines and Planes, Gaussian Distribution, Probability Density Functions

Predictive Modeling and ML

Industry Application, Core Concepts (Train & Test samples, model metrics)

Linear/ Non-linear Regression models and implementation in Scikit-learn

Classification models (Logistic, SVM) and implementation in Scikit-learn

Classification models (Decision Trees) and implementation in Scikit-learn

Ensemble Learning: Tree-based and others

Machine Learning with Python-From Linear Models to Deep Learning

Working with Linear Classifiers and Linearly Seperable Data

Estimatiing Model Parameters using Perceptron Algorithm and Gradient Descent

Working with Linear SVM

Linear Regression

Non Linear Models and Feature Maps

SVM with Kernels

Recommendation Engines using Memory Based and Model Based Methods (Matrix Factorization)

Introduction to Neural Networks: Activation Functions, Forward pass, Backward pass

Recurrent Neural Networks and Sequence Models

Convolutional Neural Networks, Using Pytorch for Neural Network Models Implementation

KMeans Clustering and EM Algorithm

Gaussian Mixture Models for Collaborative Filtering

Reinforcement Learning

Statistics Basics

Statistics Concepts - Descriptive Statistics (mean, median, variance, std. dev, percentiles)

Understanding Univariate and Multivariate Distributions through plots (Histograms, Bar Plots, Box Plots, Two-way Tables, Scatter Plots, Q-Q Plots)

Correlation, Inferential Statistics (Point & Interval Estimation and use of various statistics)

CLT and Law of large numbers

Foundations of Statistics

Parametric Statistical Models, Parametric Estimation and Confidence Interval

Delta Method and Confidence Intervals

Introduction to Hypothesis Testing, and Type 1 and Type 2 Errors

Total Variation Distance, Kullback-Leibler (KL) Divergence, and the Maximum Likelihood Principle, MLE

Covariance Matrices, Multivariate Statistics, and Fisher Information

Maximum Likelihood Estimation (Continued) and the Method of Moments, M Estimation

Hypothesis Testing: ?2 distribution and T-test, Hypothesis Testing: Wald’s test, Likelihood Ratio Test, and Implicit Hypothesis

Hypothesis Testing: ?2-test for Multinomial Distribution, Goodness of Fit Test; Hypothesis Testing: Kolmogorov-Smirnov Test, Kolmogorov Lilliefors Test, QQ-plot

Introduction to Bayesian Statistics; Jeffrey’s Prior and Bayesian Confidence

Linear Regression 1; Linear Regression 2

Introduction to Generalized Linear Model: Exponential Families; The Canonical Link Function

Deep Learning

Basic Text Processing and NLP: Using Regex, creating tfidf features, POS Tagging and dependency parsing

DL in Practice: Using tf/pytorch to build simple neural networks, understand automatic differentiation, carry out gradient computations

DL in NLP 1: LSTMs and GRUs, Encoder Decoder Architecture for Translation

DL in NLP 2: BERT based models

DL In Computer Vision1: Use transfer learning to build image classifiers. Build multiclass and multilabel classifiers

DL In Computer Vision2: Single Shot Object Detection, measuring Object Detector Performance, custom labelling and custom training