AI & ML Certification Course

Data Preprocessing

• Introduction to Data cleaning
• Categorical Variable and Variable Level Exploration
• Data Treatment
• Lab: Data Preprocessing

Principal Component Analysis

• Principal Component Analysis

Learning Outcomes

Explore the basics of data pre-processing along with an introduction to Principal Component Analysis (PCA). Identify categorical variables and explore variable levels to grasp their impact on data analysis.

Introduction to Machine Learning

• Supervised and Unsupervised Learning
• Classification: K-Nearest Neighbor (KNN) Classifier
• Lab: Building Classification Models Using KNN

Bayes Classifier

• Bayes Classifier using Unimodal and Multimodal Distributions
• Lab: Building Bayes Classifier Using Unimodal and Multimodal Distributions

Discriminative Learning Methods for Classification

• Logistic Regression and Perceptron
• Lab: Building Classifiers using Logistic Regression and Perceptron
• Support Vector Machines (SVMs)
• Lab: Building classification models using SVM

Regression: Linear Model for Regression

• Simple Linear Regression and Multiple Linear Regression
• Lab: Building Regression Models using simple and multiple linear regression
• Nonlinear Regression using Polynomial Curve fitting and Polynomial Regression
• Lab: Building Regression Models using Polynomial Curve Fitting and Polynomial Regression

Time Series Prediction

• Autoregression (AR) Model
• Lab: Time Series Data Analysis and Prediction using Autoregression
• Moving Average (MA) Model and ARIMA model
• Lab: Time Series Data Analysis and Prediction using MA and ARIMA Models

Learning Outcomes

Explore machine learning basics, distinguishing between supervised and unsupervised learning: Master K-Nearest Neighbour (KNN) classification, Bayes classifier, and regression. Develop skills in Time Series Prediction using AR, MA, and ARIMA models.

Decision Tree

• Decision Tree Foundations
• Algorithm
• Validation and Overfitting
• Pruning
• Tree Building Model Selection
• Instance-based Learning and Their Application to Recommender Systems
• Clustering and Holt-Winters

Cluster Analysis

• Algorithm K-Means
• Unsupervised Learning
• Time series using Holt-Winters
• Introduction to Techniques for Class Imbalanced Problems.

Learning Outcomes

Discover Decision Trees, Clustering with K-means, and techniques to address Class Imbalance. Learn about pruning, Tree Building Model Selection, and handling Recommender systems.

Deep Learning for AI

Introduction to Deep Learning and its Concepts

• ANN
• ReLu
• Hyperparameters
• Weight Sharing and Activation
• Lab: FC
• Fine Tuning
• Gradient Descent Model
• CNN Hyperparameters
• Prediction of Probability of Defaults with ANNs
• Lab: CNN I Modelling
• Deep Learning Essentials
• TensorFlow
• Keras
• Linear Regression and Model Building on TensorFlow
• CNN
• Difference Between ANN and CNN
• Lab: Fine Tuning
• Fully Connected Networks and its Computations
• Gradient Descent
• Regularisation and Convolusions
• Lab: FCN
• RNN
• Vanilla NN
• RNN Process Sequence
• Sequential Processing of Non-processing Data
• RNN Computational Graphs with Variables
• Language sampling model
• Lab: Building an RNN Model
• Working with Filters and Layers in ANN
• Convolution Layer
• Kernal Matrix
• LSTM Models, Issues and Representations
• Lab: LSTM Gates
• CNN Architecture and Models
• RNN sequential Models
• Back Propagation in RNN
• Vanishing Gradients
• Lab: CNN and FC (RNN)
• GAN and VAE
• Problems with L2 and MSE Loss
• Gaussian Distribution
• Lab: GAN

AI for Natural Language Processing

• Introduction to LLM
• Word2Vec
• RNN
• LSTM
• GRU
• Seq2Seq
• Self-Attention Models
• Transformer
• Bert

Learning Outcomes

Explore Deep Learning foundations, covering ANN, ReLu, and advanced topics like RNNs, LSTMs, CNN architectures, and GANs. Gain a comprehensive understanding of AI in Natural Language Processing.

Image Enhancement and Restoration

• Image Enhancement Definition and Concept
• Domains and Techniques
• Power Law Transformation
• Histogram Processing
• Its Types and Methodology
• Lab: Histogram Equalisation
• Image Smoothening
• Mean Filtering
• Medial Filtering
• Spatial Filtering
• Convolution Theorem
• Gaussian Filtering and Inverse Filtering
• Edge Detection
• Lab: Gaussian Filtering
• Lab: Inverse Filtering
• Lab: Edge Detection

Image-to-Image Transformations

• Introduction to Image-to-image Transformations through Techniques like Super Resolution, Image Inpainting, Dehazing and Colorisation
• Lab: Image Colorisation

Learning Outcomes

Learn Image Enhancement techniques, including Power Law transformation, Histogram processing, and Image Smoothing. Explore Image-to-Image Transformations, covering Super-resolution, image inpainting, dehazing, and colorization for effective image restoration.

AI for Speech Processing

• Introduction to Speech Processing and Different Applications, Features to Represent the Speech Data
• Case Study: Speaker Recognition
• Lab Associated with the Theory Covered
• Case Study: Designing GMM-based Classifiers and Neural Network-based Classifiers for Speaker Recognition

Learning Outcomes

Gain insights into speech processing and its diverse applications, and explore features to represent speech data. Engage in a case study on Speaker Recognition with a corresponding hands-on lab. Further, delve into the design of GMM-based and neural network-based classifiers for Speaker Recognition, solidifying your understanding through practical application in a case study.

Low-level and High-level Segmentation

• Introduction to Image Segmentation Including Colour-based Segmentation, Texture-based Segmentation, Intensity-based Segmentation
• K-Means Clustering
• SLIC Superpixels
• Lab: K-Means

3D Computer Vision

• Introduction to Stereo Vision
• Depth Cue
• Simple Stereo Vision (2D and 3D)
• Stereo Derivations, Disparity Estimations
• Disparity Maps
• Image to Image transformation through Depth Cues
• Lab: Disparity

Segmentation and 3D Computer

• Introduction to Semantic Segmentation
• Need an Application for SS
• Learning-based Semantic Segmentation using FCN
• SegNet Decoders
• Lab: Semantic Segmentation

Registration, Motion Estimation, Tracking

• Introduction to Image Registration and its Applications
• Motion Estimation
• Types of Geometric Transformations
• Tracking and Methods of Tracking including Harris Corner Detection and KLT tTacker
• Lab: KLT Tracker

Applications of Computer Vision

• Remote Sensing
• Visual Quality Assessment
• Other Applications

Learning Outcomes

• Understand Image Segmentation, covering color, texture, and intensity-based methods. Explore K-means Clustering and SLIC Superpixels through hands-on labs.
• Introduce Stereo Vision, depth cues, and simple stereo vision (2D and 3D). Learn stereo derivations, disparity estimations, and image-to-image transformations using depth cues. Apply knowledge in the lab focused on Disparity.
• Explore semantic segmentation, its necessity, and its applications. Implement learning-based semantic segmentation using FCN and Segnet decoders in the lab.
• Learn Image Registration, motion estimation, and types of geometric transformations. Understand tracking methods, including Harris corner detection and KLT tracker through a practical lab.
• Discover the applications of computer vision in remote sensing, visual quality assessment, and other domains.

AI in Healthcare

• Computer Vision for Biomedical Problems
• Some Standard Tasks in Medical Image Analysis, Forensics, Medical Document Analysis, etc.

Applications of AI in Healthcare

• Other Healthcare and Medical Applications of ML/CV

Learning Outcomes

• Discover various ML/CV applications in healthcare, expanding your understanding of the broader impact of AI in the medical field.
• Explore Computer Vision's role in biomedical problem-solving and standard tasks in medical image analysis, forensics, and medical document analysis.