RTMNU · Final Year B.Tech Information Technology · CBCS
Deep Learning
BTIT702T.1 · Elective IV · Semester VII · 3 credits · College 30 + university 70
One of the Elective IV options.
Unit 1: Basics of Deep Learning
6 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
- History and success stories of deep learning
- McCulloch–Pitts neuron and perceptron learning
- Multilayer perceptron and backpropagation
Unit 2: Training of Feed-forward Neural Networks
6 hoursThe representation power of feed-forward networks and how they are trained: gradient descent, momentum and Nesterov accelerated gradient, stochastic gradient descent, AdaGrad, RMSProp and Adam.
Examined in the 70-mark end-semester paper
- Multilayer perceptron and backpropagation
- Gradient descent and optimisers
Unit 3: Optimisation, Activation, Initialisation and Regularisation
6 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
- Gradient descent and optimisers
- Activation functions
- Weight initialisation: Xavier and He
- Regularisation: dropout, L1 and L2
- Dataset augmentation, noise injection and ensembles
- Bias–variance trade-off and responsible AI
- Cross-validation and hyperparameter tuning
Unit 4: Convolutional Neural Networks
6 hoursThe convolution operation and pooling; LeNet, AlexNet, ZF-Net, VGGNet, GoogLeNet and ResNet; visualising convolutional networks and guided backpropagation.
Examined in the 70-mark end-semester paper
- Convolution and parameter sharing
- Pooling
- CNN architecture
- AlexNet
- CNN architectures: ZF-Net, VGGNet, GoogLeNet and ResNet
- Visualising and fooling CNNs: guided backpropagation and Deep Dream
- CNN architectures: LeNet to ResNet
Unit 5: Recurrent Neural Networks
6 hoursRecurrent neural networks and backpropagation through time; vanishing and exploding gradients; long short-term memory cells and gated recurrent units.
Examined in the 70-mark end-semester paper
- Recurrent neural networks
- Backpropagation through time, vanishing and exploding gradients
- LSTM networks
- LSTM and GRU
Unit contents are summarised from the official RTMNU syllabus . AiCanCode.org is not affiliated with Rashtrasant Tukadoji Maharaj Nagpur University (RTMNU).