KTU 2024 Scheme · Semester 6 · Program Core — Artificial Intelligence and Data Science
Deep Learning (PCADT601) Syllabus
Official KTU 2024 Scheme syllabus for Deep Learning, Semester 6, Program Core — Artificial Intelligence and Data Science (Artificial Intelligence and Data Science).
Course Code
PCADT601
Credits
4
Teaching Hours
3:1:0:0 (L:T:P:R)
CIE Marks
40
ESE Marks
60
Exam Duration
2 Hrs 30 Min
Prerequisites
PCCST503 (Machine Learning)
Semester
Semester 6
Course Objective
To get an insight into the design parameters of a deep learning model and to introduce deep learning architectures for domains such as text, multimedia and GenAI tools.
Module-wise Syllabus
Module 1
10 contact hoursIntroduction to Deep Learning: deep feed forward networks; activation functions — sign, sigmoid, tanh, ReLU, leaky ReLU, hard tanh, softmax; loss functions; training a neural network with backpropagation; practical issues in neural network training — overfitting, underfitting, hyperparameters and validation sets, estimators — bias and variance.
Module 2
12 contact hoursNetwork Design Parameters: setup and initialization — Kaiming and Xavier weight initialization, vanishing and exploding gradient problems. Optimization techniques — Gradient Descent (GD), stochastic GD, GD with momentum, GD with Nesterov momentum; parameter-specific learning rates — AdaGrad, RMSProp, Adam. Regularization techniques — L1 and L2 regularization, early stopping, dataset augmentation, parameter tying and sharing, ensemble methods, dropout, batch normalization.
Module 3
11 contact hoursConvolutional Neural Networks: basic structure of a CNN; basic layers and operations — convolution operation, effect of stride and padding, fully connected layers, CNN layers; building and training a CNN model; estimation of tensor size and number of features in CNN layers; transfer learning (size-similarity matrix); pre-trained architectures — AlexNet, ResNet-50, GoogLeNet.
Module 4
11 contact hoursDeep learning models for text processing: Recurrent Neural Network architecture; variants — deep RNN, recursive neural network, bidirectional RNN, encoder-decoder architecture, LSTM, GRU. Autoencoders and generative models: variational autoencoder, undercomplete autoencoder, stochastic encoder, denoising encoder, applications of autoencoders; generative models — Boltzmann machines, Deep Belief Networks, Generative Adversarial Networks.
Course Outcomes
- CO1Outline the standard regularization and optimization techniques for the effective training of deep neural networks.
- CO2Use Convolutional Neural Network (CNN) models and pretrained networks for different use cases.
- CO3Apply the concepts of Recurrent Neural Networks and their variants, and become familiar with natural language processing fundamentals.
- CO4Apply the concepts of autoencoders and generative models for advanced AI operations.
- CO5Apply the concept of generative models for advanced AI operations.
Assessment Pattern (CIE: 40 marks, ESE: 60 marks)
Continuous Internal Evaluation (CIE)
| Attendance | 5 |
| Assignment / Microproject | 15 |
| Internal Examination 1 (Written) | 10 |
| Internal Examination 2 (Written) | 10 |
End Semester Examination (ESE)
Total 60 marks, 2 Hrs 30 Min. See the official KTU syllabus document for the exact Part A / Part B question pattern for this course.
Textbooks & Reference Books
Textbooks
- Deep Learning — Ian Goodfellow, Yoshua Bengio, Aaron Courville (MIT Press, 1st edition, 2016)
- Neural Networks and Deep Learning — Charu C. Aggarwal (Springer International, 1st edition, 2018)
Reference Books
- Deep Learning: Core Concepts, Methods and Applications — M. Gopal (Pearson Education, 1st edition, 2022)
- Fundamentals of Deep Learning: Designing Next-Generation Machine Intelligence Algorithms — Nikhil Buduma, Nicholas Locascio (O'Reilly Media, 1st edition, 2017)
Frequently Asked Questions
How many credits is KTU Deep Learning (PCADT601)?
4 credits, with 3:1:0:0 (L:T:P:R) teaching hours per week, under the KTU 2024 Scheme.
How many modules are in the PCADT601 syllabus?
4 modules, 44 total contact hours.
What is the CIE and ESE mark split for this course?
CIE (Continuous Internal Evaluation): 40 marks. ESE (End Semester Examination): 60 marks, 2 Hrs 30 Min. Total: 100 marks.
What are the recommended textbooks for PCADT601?
Deep Learning (Ian Goodfellow, Yoshua Bengio, Aaron Courville); Neural Networks and Deep Learning (Charu C. Aggarwal).
Is this syllabus specific to one branch, or common to others too?
This Semester 6 course is listed under Program Core — Artificial Intelligence and Data Science at KTU under the 2024 Scheme — check the course header above for which branches it's common to.
Need help with KTU Deep Learning?
Learnizo offers live, 1-on-1 online tuition for KTU AI&DS subjects — matched to your exact module and semester.