Skip to content

Latest commit

 

History

History
273 lines (212 loc) · 5.65 KB

Lectures.md

File metadata and controls

273 lines (212 loc) · 5.65 KB

Machine Learning & Deep Learning

Introduction - Mathematics

Lecture 3 - 1 Hours

  • Histograms
  • Probability distributions
  • Normal distribution
  • Population parameters
  • Estimated parameters
  • Mean, variance and standard deviation

Modelling - Machine Learning - K Means Clustering

Lecture 4 - Theory + Practical - 1 Hours

  • Theory
    • K means clustering
  • Practicals
    • Algorithm
    • Elbow method

Modelling - Machine Learning - Decision Tree

Lecture 4 - Theory + Practical - 1 Hours

  • Theory
    • Decision tree classifier
  • Practicals
    • Decision tree classifier
    • Hyperparameter selection

Modelling - Machine Learning - Random Forest

Lecture 5 - Theory + Practical - 1 Hours

  • Theory
    • Random forest classifier
  • Practicals
    • Random forest classifier
    • Hyperparameter selection

Modelling - Machine Learning - Support Vector Machine

Lecture 6 - Theory + Practical - 1 Hours

  • Theory
    • Support Vector Machine
  • Practicals
    • Support Vector Machine

Modelling - Machine Learning - Principal Component Analysis

Lecture 7 - Theory + Practical - 1 Hours

  • Theory
    • Principal Component Analysis
  • Practicals
    • Principal Component Analysis

Modelling - Machine Learning - Naive Bayes Classifier

Lecture 7 - Theory + Practical - 1 Hours

  • Theory
    • Probability
    • Baye's rule
    • Naive Bayes
    • Laplacian smoothing
  • Practicals
    • Naive Bayes Classifier

Modelling - Artificial Neural Network

Lecture 8 - Theory - 1 Hour

  • What is Artificial Neural Network (ANN) ?
  • ANN Classification
  • Linear regression
  • Logistic regression
  • Gradient descent
  • Gradient descent for logistic regression
  • Back propagation for neural network
  • Activation functions

Lecture 8 - Practical - 1 Hour

  • Classification model
  • Regression model

Modelling - Convolutional Neural Network

Lecture 9 - Theory - 1 Hour

  • CNN applications
  • Convolution
  • Image convolution
  • Convolution layer
  • Pooling layer
  • ANN v/s CNN

Lecture 9 - Practical - 1 Hour

  • Image classification
  • Autoencoder - Denoise documents

Modelling - CNN Architectures

Lecture 10 - 2 Hours

  • VGG network
  • Residual network
  • Network in network
  • MobileNet

Modelling - Recurrent Neural Network

Lecture 11 - RNN - 1 Hour

  • RNN applications
  • RNN model
  • Back propagation through time
  • Different types of RNN

Lecture 11 - Word Embeddings - 1 Hour

  • Featurized representation
  • Visualizing word embeddings
  • Using word embeddings
  • Embedding matrix
  • Learning word embeddings

Modelling - Generative Adversarial Networks

Lecture 12 - Theory - 1 Hours

  • Generative Adversarial Networks
  • Feature engineering
  • Generator
  • Discriminator

Lecture 12 - Practical - 1 Hours

  • GAN
  • DCGAN

Modelling - Hyperparameter Search

Lecture 13 - 2 Hours

  • Parameters
  • Hyperparameters
  • Parameters v/s hyperparameters

Modelling - Neural Architecture Search

Lecture 14 - Neural Architecture Search - 2 Hours

  • Theory
  • Practical

Modelling - Reinforcement Learning

Lecture 15 - Reinforcement Learning - 2 Hours

  • Theory
  • Practical

Modelling - Self Spervised Learning

Lecture 16 - Self Spervised Learning - 1 Hour

  • Theory
  • Practical - Gray to color image conversion

Lecture 16 - Semi Supervised Learning - 1 Hour

  • Theory
  • Practical

Modelling - Graph Neural Networks

Lecture 17 - Graph Neural Networks - 2 Hours

  • Theory
  • Practical

Dataset Collection - Open Dataset

Lecture 18 - Open Dataset - Theory - 1 Hour

  • ImageNet
  • Open Images Dataset
  • Microsoft Common Objects in Context - MS COCO dataset
  • Chinese City Parking Dataset - CCPD

Lecture 18 - TensorFlow Dataset - Practical - 1 Hour

  • Kaggle
  • TensorFlow Dataset
  • Datahub.i0
  • Batch normalization
  • Instance normalization

Dataset Collection - Feature Engineering & Data Augmentation

Lecture 19 - Feature Engineering - 1/2 Hour

  • Feature selection
  • Principal Component Analysis

Lecture 19 - Data Augmentation - 1/2 Hour

  • Data Driven AI development
  • Data augmentation

Lecture 19 - 1 Hour

Model Deployment - Edge AI

Lecture 20 - Edge Computing - Theory - 1 Hour

  • CPU
  • GPU
  • TPU
  • Raspberry pi
  • NVIDIA Jetson
  • NVIDIA Xavier
  • NVIDIA Orin
  • Google Coral
  • NVIDIA TensorRT

Lecture 20 - Edge AI - Practical - 1 Hour

  • NVIDIA TensorRT - Image classification

Model Deployment - Cloud AI

Lecture 21 - Cloud AI - Practical - 1 Hour

  • REST API
  • gRPC
  • NVIDIA Triton Inference Server

Model Deployment - Model Monitoring

Lecture 21 - Model Monitoring - 1 Hour

  • Bias
  • Variance
  • Bias and variance tradeoff
  • Reduce bias
  • Reduce variance
  • Regularization
  • Dropout
  • Weight decay
  • Data augmentation

Model Deployment - Interpretable AI & Explainable AI

Lecture 22 - Interpretable AI & Explainable AI - 2 Hour

  • Theory
    • Interpretable AI
    • Explainable AI
    • Interpretable AI v/s Explainable AI
  • Practical
    • Interpretable AI
    • Explainable AI

Introduction - Machine Learning Design Patterns

Lecture 23 - Machine Learning Design Patterns - 2 Hours

  • Problem representation design patterns
    • Reframing design pattern
    • Ensables design pattern
    • Rebalancing design pattern
  • Model training design patterns
    • Transfer learning design pattern
    • Checkpoints design pattern
  • Reproducibility design patterns
    • Transform design pattern
    • Model versioning design pattern
  • Responsible AI design patterns
    • Fairness lens design pattern

Project Scoping - Applications

Lecture 24 - 1 Hour

  • Image classification
  • Object detection
  • Image segmentation
  • Image captioning

Lecture 24 - 1 Hour

  • Machine translation - LSTM
  • Machine translation - Transformer

Lecture 25 - 2 Hours