RTMNU · Third Year B.Tech Artificial Intelligence & Data Science · NEP 2020
Deep Learning
BAD6T21 · Semester VI · 3 credits · College 30 + university 70
Unit 1: Mathematical Preliminaries and Machine Learning Basics
8 hoursLinear algebra, principal component analysis, probability and statistics, numerical methods and gradient-based optimisation; training, validation and test sets, neural, convolutional and recurrent networks, backpropagation, performance metrics, hyperparameters and debugging.
Examined in the 70-mark end-semester paper
- Eigen-decomposition, PCA and SVD
- Gradient descent variants
- Cross-validation and model evaluation
- Feed-forward network architectures
- Backpropagation
- Hyperparameters in deep learning
Unit 2: Introduction to Deep Networks
7 hoursProblems with backpropagation and modern approaches, autoencoders and representation learning, regularisation, dropout and optimisation strategies.
Examined in the 70-mark end-semester paper
- Vanishing and exploding gradients
- Autoencoders and VAEs
- Regularisation and dropout
- Optimisers: AdaGrad, RMSProp and Adam
Unit 3: Sequence Learning and LSTMs
7 hoursDeep recurrent networks, bidirectional networks and encoder–decoder architectures; LSTMs and building an LSTM network.
Examined in the 70-mark end-semester paper
- Recurrent neural networks
- Bidirectional and deep RNNs
- Encoder-decoder sequence models
- LSTM
Unit 4: Applications
7 hoursA deep convolutional network for Telugu OCR and its performance analysis; LSTM networks for text processing.
Examined in the 70-mark end-semester paper
- Training a CNN
- CNN architectures: LeNet to ResNet
- LSTMs and GRUs for language
Unit 5: GANs and Latest Advances
7 hoursGenerative adversarial networks, building and training GANs, GAN variants and current results, and the limitations and weaknesses of deep learning.
Examined in the 70-mark end-semester paper
- Generative adversarial networks
- Types of GANs
- Explainable and responsible AI
Unit contents are summarised from the official RTMNU syllabus . AiCanCode.org is not affiliated with Rashtrasant Tukadoji Maharaj Nagpur University (RTMNU).