Deep Learning
24CST401 · S4 · 4 credits
Official KTU 2024 syllabus subject — study with practice, PYQs and AI tools.
Have an exam tomorrow?
Get a high-yield attack plan focused on maximum marks per minute. Zero fluff, strict PYQ priority, active recall, and trap avoidance.
Strategy: High-yield concepts + PYQs + weak areas
Why:Foundation module
Why:High exam relevance
Why:Core module
Why:Applied concepts
Why:Standard exam coverage
Module 01Introduction to Deep Learning (8 hours)expand
Introduction to deep learning, Deep feed forward network, Training deep models - introduction, setup and initialization issues, Vanishing and exploding gradient problems, Optimization techniques - Gradient Descent (GD), Stochastic GD, GD with momentum, GD with Nesterov momentum, AdaGrad, RMSProp, Adam. Regularization Techniques - L1 and L2 regularization, Early stopping, Dataset augmentation, Parameter tying and sharing, Ensemble methods, Dropout.
Module 02Convolutional Neural Networks (7 hours)expand
Convolutional Neural Networks – Convolution operation, Motivation, Pooling, CNN architecture, MLP versus CNN - Convolution and Pooling as an infinitely strong prior, Variants of convolution functions, Structured outputs, Data types, Efficient convolution algorithms. Practical use cases for CNNs, Case study - Popular CNN architecture – LeNet, AlexNet.
Module 03Recurrent neural networks (8 hours)expand
Recurrent neural networks – Sequence learning problems, Backpropagation Through Time, The Problem of vanishing and exploding gradients, Computational graphs, RNN design, encoder – decoder sequence to sequence architectures, deep recurrent networks, recursive neural networks, modern RNNs LSTM and GRU.
Module 04Graphical Models (6 hours)expand
Graphical models – Core concepts in graphical models- Conditional Independence. Joint and Marginal Probabilities, Factorization and Representation of Distributions. Types of graphical models-Bayesian Networks, Markov networks, Boltzmann Machines. Restricted Boltzmann Machines (RBMs).
Module 05Advanced Deep learning Topics (7 hours)expand
Autoencoders, Variational AutoEncoder, Generative Adversarial Networks, Auto-Regressive Networks. Introduction to attention mechanism and Transformers. Transfer Learning and Domain Adaptation.
Exam-focused notes are optional context — Continue Learning above works even without them.
Notes for this subject aren't written yet
Continue Learning above adapts to this subject anyway — it teaches from your active syllabus.