Introduction to DL and GenAI - Prof. Balaji Srinivasan , Prof. Ganapathy Krishnamurthi

1.1 Introduction to Artificial Neural Networks

1.2 Introduction to Pytorch and Tensors

1.3 Pytorch Installation and Environment Setup

1.4 Tensor - Creation Methods

1.5 Tensor - Manipulation Methods

1.6 Tensor - Reshaping and Indexing Methods

1.7 Pytorch and Numpy Integration

1.8 Reproducibility in Pytorch

1.9 Hardware Acceleration in Pytorch

1.10 Common Pytorch Errors

1.11 Linear Regression in Pytorch: Data Creation and Loading

1.12 Linear Regression in Pytorch: Data Visualization

1.13 Linear Regression in Pytorch: Model Definition

1.14 Linear Regression in Pytorch: Custom Loss Function

1.15 Linear Regression in Pytorch: Training Loop

1.16 Model Serialization in Pytorch

1.17 Neuron: Biology to Mathematics

1.18 Fundamental Tricks in Deep Learning

1.19 Feedforward Network Architecture

1.20 Feedforward Network Architecture: Hands-on Example

1.21 Activation Functions

2.1 Universal Approximation Theorem

2.2 Forward Pass and Loss Functions

2.3 Gradient Descent

2.4 Customer Churn Prediction: Overview

2.5 Customer Churn Prediction: Data Processing

2.6 Customer Churn Prediction: Neural Network Package Imports

2.7 Customer Churn Prediction: Neural Network Implementation

2.8 Customer Churn Prediction: Weight Initialization

2.9 Customer Churn Prediction: Training and Validation Setup

2.10 Customer Churn Prediction: Training Loop Implementation

2.11 Customer Churn Prediction: Model Training

2.12 Backpropagation: Simple Scalar Chain

2.13 Backpropagation: Multilayer Perceptron

2.14 The XOR Problem: Introduction

2.15 The XOR Problem: Forward Pass

2.16 The XOR Problem: Backpropagation

2.17 Fashion MNIST Classification: Overview and Data Processing

2.18 Fashion MNIST Classification: MLP Implementation in Pytorch

2.19 Fashion MNIST Classification: Forward Pass

2.20 Fashion MNIST Classification: Training Loop

2.21 Fashion MNIST Classification: Validation Loop

2.22 Fashion MNIST Classification: Model Training

3.1 Introduction to Vision Modelling

3.2 From MLP to Convolution

3.3 Filters, Feature Maps and Receptive Field

3.4 CNN: Hyperparameters

3.5 CNN- Network Layers

3.6 CNN: Architecture and Data Augmentation

3.7 Convolution from scratch

3.8 Pooling from scratch

3.9 Convolution in Pytorch

3.10 Pooling in Pytorch

3.11 CNN Image Classification: Simple CNN implementation

3.12 CNN Image Classification: Data preprocessing and Dataloaders

3.13 CNN Image Classification: Model Training

4.1 Famous CNN Architectures

4.2 Transfer Learning

4.3 Image Classification with Transfer Learning: Introduction

4.4 Image Classification with Transfer Learning: Dataset Accumulation

4.5 Image Classification with Transfer Learning: Data Exploration

4.6 Image Classification with Transfer Learning: Data Preprocessing

4.7 Image Classification with Transfer Learning: Pretrained Model Usage

4.8 Image Classification with Transfer Learning: Training and Evaluation

4.9 Image Segmentation: Introduction to U-Net

4.10 Image Segmentation: Data Accumulation

4.11 Image Segmentation: Data Preprocessing

4.12 Image Segmentation: Training and Evaluation

5.1 Optimization in DL -- Challenges

5.2 Convex Optimization

5.3 Gradient Descent Algorithm

5.4 Accelerating Gradient Descent

5.5 Gradient Descent Demo

5.6 Accelerating Gradient Descent -- Demo

6.1 Introduction to Generative Models

6.2 Introduction to Latent Space

6.3 Types of Generative Models

6.4 Introduction to Generative Adversarial Networks

6.5 Fashion MNIST Generation using GAN: Overview and Data Processing

6.6 Fashion MNIST Generation using GAN: Model Implementation

6.7 Fashion MNIST Generation using GAN: Adversarial Training Loop

6.8 Fashion MNIST Generation using GAN: Model Evaluation

7.1 Introduction to Sequence Modeling

7.2 Mathematical Modeling of Sequences

7.3 Language Modelling

7.4 Recurrent Neural Networks

7.5 Vanilla RNN : Python Demo

8.1 Memory Based RNN models

8.2 Bidirectional RNN

8.3 Encoder-Decoder Architectures

8.4 Sentiment Analysis : Demo

9.1 Word Embeddings

9.2 The Attention Mechanism

9.3 Neural Machine Translation -- Overview

9.4 Neural Machine Translation -- Code Walkthrough

10.1 From RNN Attention to Transformer Attention

10.2 Parts of the Transformer Architecture

10.3 The Original Transformer

10.4 From Transformers to LLMs

10.5 Modern Transformers in Language Models

11.1 Vision Transformers

11.2 Demo -- NLP tasks through Hugging Face and API calls

11.3 Introduction to Generative for Images

11.4 Evaluation Metrics for Image Generation

11.5 Introduction to Denoising Diffusion Probabilistic Models (DDPM)

11.7 Denoising Diffusion Implicit Models DDIM

11.6 Training DDPM

11.8 Controlled conditioned generation and Latent Diffusion

12.1 Custom Tokenization and Data preprocessing

12.2 Self-Attention Head

12.3 Multi-Head Self Attention

12.4 GPT Language Model

12.6 LLM Finetuning: Model Initialization

12.7 LLM Finetuning: Data Preparation

12.9 LLM Finetuning: Updating the weights via LoRA

12.5 Retrieval Augmented Generation(RAG)

12.8 LLM Finetuning: Data Generation via LLM