Syllabus Neural Networks - (BC05001061) Total Credits L + T + (PR/2) Assessment Pattern and Marks Total Marks C Theory Practical ESE (E) PA / CA (M) PA (I) ESE (V) 5 70 30 20 30 150 Unit No. Content 1. Introduction to Neural Networks Introduction to artificial intelligence, machine learning and neural networks, motivation for neural networks, biological neuron and artificial neuron, components of artificial neuron, neural network terminology, architecture of neural networks, types of neural networks, advantages, and limitations of neural networks. (Chapter - 1) 2. Artificial Neuron Models and Activation Functions Structure of artificial neuron models, perceptron model, single layer perceptron, multilayer perceptron, weights and bias in neural networks, activation functions including step function, sigmoid, tanh and ReLU, feedforward neural networks, representation, and interpretation of neural network models. (Chapter - 2) 3. Neural Network Learning and Training Learning rules in neural networks, supervised learning and unsupervised learning, perceptron learning rule, gradient descent optimization, forward propagation, backpropagation, training process in neural networks, overfitting and underfitting in neural networks, hyper parameters : epochs, batch size and learning rate. (Chapter - 3) 4. Neural Network Architectures Multilayer neural networks, deep feedforward networks, introduction to convolutional neural networks (CNN), convolution and pooling operations, introduction to recurrent neural networks (RNN), sequence data processing, applications of different neural network architectures. (Chapter - 4) 5. Neural Network Implementation and Applications Introduction to neural network tools and frameworks, Python libraries for neural networks, building simple neural network models using Python, training, and testing neural network models, evaluation of neural network performance, visualization of results, practical applications of neural networks in classification and prediction tasks. (Chapter - 5)
