IISc Bangalore RV University · Computer Science
Foundations of Deep Learning: Concepts and Applications
Prof. Ashwini Kodipalli, Prof. Baishali Garai, Prof. Sriram Ganapathy · 65 free video lectures
65 video lectures. Topics: Overview of Machine Learning and Deep Learning; Single Layer Perceptron; Hands-on on Single Layer Perceptron; Multi-Layer Perceptron; Activation and Loss Functions; Gradient descent algorithm; Backpropagation in Neural Network; Varients of Gradient descent and Momentum…
▶ Play on Open Lecture HallCourse page ↗
Lectures
- Overview of Machine Learning and Deep Learning
- Single Layer Perceptron
- Hands-on on Single Layer Perceptron
- Multi-Layer Perceptron
- Activation and Loss Functions
- Gradient descent algorithm
- Backpropagation in Neural Network
- Varients of Gradient descent and Momentum based techniques
- Regularization techniques in Neural Networks
- Hands-on on MLP for classification problem
- Fundamentals of Image representation and Image preprocessing and Data augmentation
- Introduction to Convolutional Neural Networks, Inspiration behind CNN, Key Components of CNN
- Convolution for RGB, Pooling Layer and Flatten Layer
- Batch Normalization
- Hands-on session on building simple CNN model
- Architecture and Implementation of AlexNet
- VGGNet (VGG16 and VGG19) and GoogleNet
- Introduction to ResNet (ResNet 34 and ResNet50)
- Introduction to Transfer Learning
- Hands-on on standard CNN architectures and transfer learning
- Hands-on on building and deep CNN ensemble model.
- Need for XAI, PCI, Occlusion method
- SHAP and LIME
- Guided backpropagation
- Grad-CAM and Class Activation Maps (CAMs)
- Hands-on implemenation of XAI methods
- Image segmentation
- U-Net with Attention and Evaluation Metric for Segmentation
- Object detection
- YOLO v1 for object detection
- Hands-on implemenation of Unet
- Creating .yaml file for the custom dataset
- Hands-on implemenation of YOLO
- Introduction to Recurrent Neural Networks
- Numerical Example for RNN
- Vanishing and exploding gradients in NN
- Backpropagation through time (BTT)
- Hands-on implementation of RNN model and its variants.
- Introduction to LSTM
- Gated Recurrent Unit
- LSTM-Encoder Decoder Structure
- LSTM with attention
- Hands-on implementation of LSTM
- Introduction to NLP
- Text Preprocessing and Text Representations
- Discrete and Distributed Word Representation
- Word Embeddings in NLP
- Hands-on implementation of RNN in NLP application
- Introduction to Autoencoders
- Types of Autoencoders
- VAE Architecture - Encoder part
- VAE Architecture - Decoder part
- Reparameterization Trick
- Hands-on implemenation of Autoencoders
- Foundations of Language Modeling and Transformers
- Transformer Architectures and Attention Mechanisms
- In-context learning and Self-Supervised Learning in LLMs
- Large Language Models: Tokenization, Generation, and Sampling
- Pre-training and Instruction Tuning of Large Language Models
- Reinforcement Learning for Aligning Large Language Models
- Reasoning, Retrieval, and Efficiency in Post-trained LLMs
- Evaluation, Benchmarking, and Impact of Foundation Models
- LLMs demo -TA
- Diffusion Models for Generative Modelling
- GANs and Diffusion models demo -TA
From NPTEL (IITs and IISc) · CC BY-NC-SA. More from IISc Bangalore RV University.
Open Lecture Hall is free, with no ads and no sign-up. It is not run by or connected to any university; the lectures are the universities' own free recordings, played from where they host them. About & rights