PG Program in Artificial Intelligence & Machine Learning: Business Applications

1 Project

7 Case Studies

Python is an essential programming language in the tool-kit of an AI & ML professional. In this course, you will learn the essentials of Python and its packages for data analysis and computing, including NumPy, SciPy, Pandas, Seaborn and Matplotlib.

• Python Programming Fundamentals

Python is a widely used high-level, interpreted programming language, having a simple, easy-to-learn syntax that highlights code readability.

This module will teach you how to work with Python syntax to executing your first code using essential Python fundamentals

• Python for Data Science - NumPy and Pandas

NumPy is a Python package for scientific computing like working with arrays, such as multidimensional array objects, derived objects (like masked arrays and matrices), etc. Pandas is a fast, powerful, flexible, and simple-to-use open-source library in Python to analyse and manipulate data.

This module will give you a deep understanding of exploring data sets using Pandas and NumPy.

• Data Visualization using Python

Data visualization is an important skill and one can create compelling visual representations of data to enable effective analysis and communication of insights. Python provides libraries to do this in a simple and effective manner.

• Exploratory Data Analysis

Exploratory Data Analysis, or EDA, is essentially a type of storytelling for statisticians. It allows us to uncover patterns and insights, often with visual methods, within data.

This module will give you a deep insight into EDA in Python and visualization tools-Matplotlib and Seaborn.

• Data Pre-processing

Data preprocessing is a crucial step in any machine learning project and involves cleaning, transforming, and organizing raw data to improve its quality and usability. The preprocessed data is used both analysis and modeling.

3 Quizzes

1 Project

Statistical Learning is a branch of applied statistics that deals with Machine Learning, emphasizing statistical models and assessment of uncertainty. This course on statistics will work as a foundation for Artificial Intelligence and Machine Learning concepts learnt in this AI ML PG program.

• Descriptive Statistics
  The study of data analysis by describing and summarising numerous data sets is called Descriptive Analysis. It can either be a sample of a region’s population or the marks achieved by 50 students.
  This module will help you understand Descriptive Statistics in Python for AI ML.
• Inferential Statistics
  Inferential Statistics helps you how to use data for estimation and assess theories. You will know how to work with Inferential Statistics using Python.
• Probability & Conditional Probability
  Probability is a mathematical tool used to study randomness, like the possibility of an event occurring in a random experiment. Conditional Probability is the likelihood of an event occurring provided that several other events have also occurred.
  In this module, you will learn about Probability and Conditional Probability in Python for AI ML.
• Hypothesis Testing
  Hypothesis Testing is a necessary Statistical Learning procedure for doing experiments based on the observed/surveyed data.
  You will learn Hypothesis Testing used for AI and ML in this module.
• Chi-square & ANOVA
  Chi-Square is a Hypothesis testing method used in Statistics, where you can measure how a model compares to actual observed/surveyed data.
  Analysis of Variance, also known as ANOVA, is a statistical technique used in AI and ML. You can split observed variance data into numerous components for additional analysis and tests using ANOVA.
  This module will teach you how to identify the significant differences between the means of two or more groups.

1 Project

10 Case Studies

Supervised Machine Learning aims to build a model that makes predictions based on evidence in the presence of uncertainty. In this course, you will learn about Supervised Learning algorithms of Linear Regression and Logistic Regression.

• Linear Regression

Linear Regression is one of the most popular supervised ML algorithms used for predictive analysis, resulting in producing the best outcomes. You can use this technique to assume a linear relationship between the independent variable and the dependent variable. You will cover all the concepts of Linear Regression in this module.

• Logistic Regression

Logistic Regression is also one of the most popular supervised ML algorithms, like Linear Regression. It is a simple classification algorithm where you can predict the categorical dependent variables with independent variables’ assistance. You will cover all the concepts of Logistic Regression in this module.

• Decision Trees

A decision tree is a Supervised ML algorithm, which is used for both classification and regression problems. It is a hierarchical structure where internal nodes indicate the dataset features, branches represent the decision rules, and each leaf node indicates the result.

1 Project

8 Case Studies

Ensemble methods help to improve the predictive performance of Machine Learning models. In this machine learning online course, you will learn about different Ensemble methods that combine several Machine Learning techniques into one predictive model in order to decrease variance, bias or improve predictions.

• Bagging and Random Forests

In this module, you will learn Random Forest, a popular supervised ML algorithm that comprises several decision trees on the provided several subsets of datasets and calculates the average for enhancing the predictive accuracy of the dataset, and Bagging, an essential Ensemble Method.

• Boosting

Boosting is an Ensemble Method which can enhance the stability and accuracy of machine learning algorithms, converting them into robust classification, etc.

• Cross Validation

Cross-validation is a technique used to evaluate the performance of machine learning models, which helps in getting a clearer picture of an ML model's generalization ability. K-fold cross validation is one of the most common approaches.

• Class Imbalance Handling

Class imbalance handling is a crucial task in machine learning when the distribution of classes in the dataset is uneven. There are different techniques like RandomUnderSampler, SMOTE, etc. used for this purpose.

• Hyperparameter Tuning

Hyperparameter tuning is the process of finding the optimal values for hyperparameters in a machine learning algorithm. It involves exploring different combinations of hyperparameter values and evaluating their impact on the model's performance, often using techniques like grid search, random search, etc

1 Project

4 Case Studies

Unsupervised Learning finds hidden patterns or intrinsic structures in data. In this machine learning online course, you will learn about commonly-used clustering techniques like K-Means Clustering and Hierarchical Clustering along with Dimension Reduction techniques like Principal Component Analysis.

• K-Means Clustering

K-means clustering is a popular unsupervised ML algorithm, which is used for resolving the clustering problems in Machine Learning. In this module, you will learn how the algorithm works and later implement it. This module will teach you the working of the algorithm and its implementation.

• Hierarchical Clustering

Hierarchical Clustering is another popular unsupervised ML technique or algorithm, which is used for building a hierarchy or tree-like structure of clusters. For example, you can combine a list of unlabeled datasets into a cluster in the hierarchical structure.

• Principal Component Analysis (PCA)

PCA is a dimensionality reduction technique used to transform a high-dimensional dataset into a lower-dimensional space. This can help in choosing and retaining only the variable which capture the highest amount of variablity in the data as they will be the most important ones.

This last module of the machine learning online course will discuss the model deployment techniques and techniques around making your model scalable, robust, and reproducible.

• Packaging Models
  Model Packaging helps you package all the necessary assets to host a model as a web service. It also enables you to download either a fully built Docker image or the files required to make one.
• Rest APIs, Dockers
  RESTful API, also known as Representational State Transfer, is an API that uses HTTP requests like GET, PUT, POST, and DELETE to communicate with web services.
  Docker is one of the most popular tools, which is used to create, deploy, and run applications with the help of containers.
  This module will teach how to package up an application using containers.
• ML Pipeline and Model Scalability
  In this module, you will learn everything you need to know about ML Pipeline and Model Scalability used for ML models.

1 Project

4 Case Studies

In this artificial intelligence online course, you will learn about the basic building blocks of Artificial Neural Networks. In this deep learning online course, you’ll learn how Deep Learning Networks can be successfully applied to data for knowledge discovery, knowledge application, and knowledge-based prediction.

• Deep Learning and its history

Deep Learning carries out the Machine Learning process using an ‘Artificial Neural Net’, which is composed of several levels arranged in a hierarchy. It has a rich history that can be traced back to the 1940s, but significant advancements occurred in the 2000s with the introduction of deep neural networks and the availability of large datasets and computational power.

• Multi-layer Perceptron

The multilayer perceptron (MLP) is a type of artificial neural network with multiple layers of interconnected neurons, including an input layer, one or more hidden layers, and an output layer. It is a versatile architecture capable of learning complex patterns from data.

• Activation functions

Activation Function is used for defining the output of a neural network from numerous inputs.

• Backpropagation

Backpropagation is a key algorithm used in training artificial neural networks, enabling the calculation of gradients and the adjustment of weights and biases to iteratively improve the performance of a neural network.

• Optimizers and its types

Optimizers are algorithms used to adjust the parameters of a neural network model during training to minimize the loss function. Different types of optimizers are Gradient Descent, RMSProp, Adam, etc.

• Weight Initialization and Regularization

Weight initialization is the process of setting initial values for the weights of a neural network, which can significantly impact the model's training and convergence. Regularization is a technique used in machine learning/ neural networks to prevent the model from overfitting, which helps improve the model's generalization ability.

1 Project

4 Case Studies

The next module is the Computer Vision course. This module will reflect on the ability of a computer system to see and make sense of visuals using CNN (Convolutional Neural Network). It will enable you to efficiently handle image data for the purpose of feeding into CNNs.

• Overview of Computer Vision (CV)

In this module, you will drive through all the business applications of computer vision and learn how it impacted several business industries.

• Understanding images (Color Pixel Theory and Image Representation)

This module will teach you how to process the image and extract all the data from it, where you can use the data for image recognition in deep learning.

• Convolution Operation and Convolutional Neural Networks (CNNs)

Convolutional Neural Networks (CNN) are used for image processing, classification, segmentation, and numerous other applications. This module will give you a deep understanding of CNNs from scratch.

• CNN Architectures

CNN architectures are specialized deep learning models designed for processing grid-like data, such as images or audio. Common architectures include VGG16, ResNet, InceptionNet, etc.

• Transfer Learning

Transfer learning is a technique in neural networks where a pre-trained model, usually trained on a large dataset, is used as a starting point for solving a related task. It allows leveraging the knowledge and learned features from the pre-trained model, reducing the need for extensive training on limited data, and often leads to improved performance and faster convergence.

1 Project

3 Case Studies

This module of the NLP online course talks about yet another interesting implementation of Neural Networks that revolves around equipping computers to understand human language. You will learn to understand sentiments from the texts.

• Overview of Natural Language Processing (NLP)

This module will get you comfortable with the introduction to NLP and, later, teach you all the essential business applications you need to know about NLP. Natural Language Processing (NLP) applies computational linguistics to build real-world applications, which work with languages consisting of varying structures. First, we teach the computer to learn languages and then expect it to understand them with relevant, efficient algorithms.

• Text Preprocessing

Text preprocessing is a technique for cleaning and preparing text data. In this module, you learn how to work with text data using all the steps involved in preprocessing a text, such as Text Cleansing, Tokenization, Stemming, etc.

• Word Representation and Vectorization Techniques

Word representation and vectorization techniques aim to convert words or text into numerical representations that can be processed by machine learning models. Methods like one-hot encoding and word embeddings (such as Word2Vec or GloVe) provide effective ways to capture semantic and contextual information about words.

• Recurrent Neural Networks (RNNs)

Recurrent Neural Networks (RNNs) are a type of neural network architecture specifically designed to process sequential data, such as time series or natural language. They introduce feedback connections and excel in capturing temporal dependencies in data.

• Long Short Term Memory Networks (LSTMs)

Long Short-Term Memory (LSTM) networks are a specialized type of recurrent neural network (RNN) designed to overcome the limitations of standard RNNs. They utilize memory cells and a gating mechanism which helps in capturing long-term dependencies.

• Sentimental Analysis

Sentiment Analysis is an NLP technique to check whether the data is positive, negative, or neutral. Twitter is the most commonly used example of Sentiment Analysis.

Gain an understanding of what ChatGPT is and how it works, as well as delve into the implications of ChatGPT for work, business, and education. Additionally, learn about prompt engineering and how it can be used to fine-tune outputs for specific use cases.

• Overview of ChatGPT and OpenAI
• Timeline of NLP and Generative AI
• Frameworks for understanding ChatGPT and Generative AI
• Implications for work, business, and education
• Output modalities and limitations
• Business roles to leverage ChatGPT
• Prompt engineering for fine-tuning outputs
• Practical demonstration and bonus section on RLHF

Dive into the development stack of ChatGPT by learning the mathematical fundamentals that underlie generative AI. Further, learn about transformer models and how they are used in generative AI for natural language.

• Mathematical Fundamentals for Generative AI
• VAEs: First Generative Neural Networks
• GANs: Photorealistic Image Generation
• Conditional GANs and Stable Diffusion: Control & Improvement in Image Generation
• Transformer Models: Generative AI for Natural Language
• ChatGPT: Conversational Generative AI
• Hands-on ChatGPT Prototype Creation
• Next Steps for Further Learning and understanding

1 Project

The last module in this Artificial Intelligence and Machine Learning online course is Recommendation Systems. A large number of companies use recommender systems, which are software that select products to recommend to individual customers. In this course, you will learn how to produce successful recommender systems that use past product purchase and satisfaction data to make high-quality personalized recommendations.

• Popularity-based Model
  A popularity-based model is a recommendation system, which operates based on popularity or any currently trending models.
• Market Basket Analysis
  Market Basket Analysis, also called Affinity Analysis, is a modeling technique based on the theory that if you purchase a specific group of items, then you are more probable to buy another group of items.
• Content-based Model
  First, we accumulate the data explicitly or implicitly from the user. Next, we create a user profile dependent on this data, which is later used for user suggestions. The user gives us more information or takes more recommendation-based actions, which subsequently enhances the accuracy of the system. This technique is called a Content-based Recommendation System.
• Collaborative Filtering
  Collaborative Filtering is a collective usage of algorithms where there are numerous strategies for identifying similar users or items to suggest the best recommendations.
• Hybrid Recommendation Systems
  A Hybrid Recommendation system is a combination of numerous classification models and clustering techniques. This module will lecture you on how to work with a Hybrid Recommendation system.