RTMNU · Third Year B.Tech Computer Science & Engineering (Data Science) · NEP 2020
Deep Learning
BDS6T22.3 · Elective II · Semester VI · 3 credits · College 30 + university 70
One of the Elective II options.
Unit 1: Basics of Deep Learning
7 hoursThe history of deep learning; the McCulloch–Pitts neuron and thresholding logic; perceptrons, the perceptron learning algorithm and its convergence; multilayer perceptrons and their representation power; sigmoid neurons and feed-forward networks.
Examined in the 70-mark end-semester paper
- McCulloch–Pitts neuron and thresholding logic
- Perceptron learning rule
- Multilayer perceptron
- Feed-forward network architectures
Unit 2: Training of Feed-forward Neural Networks
8 hoursThe representation power of feed-forward networks and their training; gradient descent, momentum-based and Nesterov accelerated gradient descent, stochastic gradient descent, AdaGrad, RMSProp and Adam.
Examined in the 70-mark end-semester paper
- Backpropagation
- Gradient descent variants
- Momentum and Nesterov accelerated gradient descent
- Optimisers: AdaGrad, RMSProp and Adam
Unit 3: Optimization, Activation, Initialization and Regularization
7 hoursThe gradient-descent family again; sigmoid, tanh and ReLU activations; Xavier and He initialisation; bias and variance, overfitting and hyperparameter tuning; L1 and L2 regularisation, data augmentation, early stopping and parameter sharing and tying.
Examined in the 70-mark end-semester paper
- Activation functions
- Weight initialisation: Xavier and He
- Bias-variance trade-off
- Hyperparameters in deep learning
- Regularisation and dropout
- Data augmentation
- Early stopping, parameter sharing and ensembles
Unit 4: Convolutional Neural Networks
7 hoursThe convolution operation and pooling; LeNet, AlexNet, ZF-Net, VGGNet, GoogLeNet and ResNet; visualising CNNs and guided backpropagation.
Examined in the 70-mark end-semester paper
- Convolution and feature maps
- Padding, stride and pooling
- CNN architectures: LeNet to ResNet
- Visualising CNNs: guided backpropagation and Deep Dream
Unit 5: Recurrent Neural Networks
7 hoursRecurrent neural networks and backpropagation through time; vanishing and exploding gradients; LSTM cells and gated recurrent units.
Examined in the 70-mark end-semester paper
- Recurrent neural networks
- Backpropagation through time
- Vanishing and exploding gradients
- LSTM
- GRU
Unit contents are summarised from the official RTMNU syllabus . AiCanCode.org is not affiliated with Rashtrasant Tukadoji Maharaj Nagpur University (RTMNU).