AI5006 - Deep Learning

Download as pdf or txt
Download as pdf or txt
You are on page 1of 6

Uka Tarsadia University

B. Tech.
Semester VI

DEEP LEARNING

AI5006

EFFECTIVE FROM July-2022


Syllabus version: 1.00
Teaching Scheme
Subject
Subject Title Hours Credits
Code
Theory Practical Theory Practical
AI5006 Deep Learning 3 2 3 1

Theory Practical
Subject Examination Examination Total
Subject Title
Code Marks Marks Marks
Internal External CIE
AI5006 Deep Learning 40 60 50 150

Objective of the course:


• To introduce various machine learning and deep learning techniques.
• To understand the effect of various hyperparameters on model’s performance.
• To develop Convolution Neural Network and Recurrent Neural Network from the
scratch and use of these learning algorithms for various applications.
• To learn the various Auto Encoders and Generative Adversarial Network and its use for
various applications.

Course outcomes:
Upon completion of the course, the student shall be able to
CO1: Understand machine learning and deep learning.
CO2: Identify difference between Shallow and deep neural network.
CO3: Use different hyperparameters for regulation and optimization.
CO4: Develop the Convolution Neural Network.
CO5: Understand Recurrent Neural Network for sequential pattern learning.
CO6: Learn various Auto Encoders and Generative Adversarial Network.

Sr. No. Topics Hours


Unit – I
1 Introduction to Machine Learning: 6
Machine learning, Applications and Future scope of machine
learning, Types of learning – Supervised, unsupervised and
reinforcement, Training versus Testing, Logistic regression,
Multilayer perceptron model.
Unit – II
2 Shallow and Deep Neural Networks: 8
Introduction to neural network, Activation functions, Gradient
descent for neural networks, Introduction to deep learning,
Machine learning vs Deep learning, Deep L-layer neural network,
Forward propagation in deep network, Matrix dimension, Deep
representation, Building blocks of deep neural network, Forward
and backward propagation.
Unit – III
3 Deep Neural Networks - Hyper Parameter Tuning, 8
Regularization and Optimization:
Training, validation and testing datasets, Bias and variance,
Regularization, Dropout, Regularization methods, Normalizing
input, Vanishing gradient, Weight initialization of deep neural
network, Numerical approximation of gradients, Gradient checking,
Mini-batch gradient descent, Exponentially weighted averages,
Gradient descent with momentum, RMSprop, Adam optimization
algorithm, Learning rate decay, The problem of local optima,
Optimizers, Tuning process, Appropriate scale to pick
hyperparameters, Hyperparameters tuning, Normalizing
activations in a network, Fitting batch norm into a neural network,
Softmax Regression, Training a softmax classifier, Deep learning
frameworks.
Unit – IV
4 Convolutional Neural Network: 8
Introduction to convolution operation, 1-D and 2-D convolution,
Importance of convolution, One Layer of a convolutional neural
network (CNN), Convolutional layer, Pooling layers, Filters and
feature maps, Activation functions, Flattening layer, Fully
connected layer, Data augmentation, Padding, Stride, Convolutional
neural network, Learning by a model, Training the network,
Advantages of hierarchical features, Binary classification using
CNN, Multi class classification, CNN on real images, Applications.
Unit – V
5 Sequence Models - Recurrent Neural Networks: 8
Sequence models, Notation, Recurrent Neural Network Model,
Backpropagation through time, Different types of RNNs, Language
model and sequence generation, Sampling novel sequences,
Vanishing gradients with RNNs, Gated Recurrent Unit (GRU), Long
Short Term Memory (LSTM), Bidirectional RNN, Deep RNNs.
Unit – VI
6 Advance Topics: 7
Introduction to autoencoders, Overview of undercomplete
autoencoders, Sparse autoencoders, Denoising autoencoders,
Variational autoencoders, Introduction to Generative models,
Generative Adversarial Networks (GANs).
Sr. No. Deep Learning (Practical) Hours
1. Introduction and installation of Tensorflow. 4
2. Perform implementation of logistic regression model with 2
tensorboard to visualize computational graphs and learning.
3. Perform implementation of multilayer model for handwritten 2
digits in tensorflow by using MNIST dataset.
4. Perform implemention various optimizers using tensorflow. 4
5. Give an introduction of keras. Also apply preprocessing using 4
keras.
6. Write a program to implement layer activation function using 2
keras and tensorflow.
7. Write a program to implement image classification using 2
Convolution Neural network.
8. Write a program to implement image classification using transfer 2
learning.
9. Implement recurrent Neural network. 2
10. Implement Generative Adversarial Network using keras. 2
11. Implementation of miniproject. 4

Text book:
1. Charu C. Aggarwal– “Neural Networks and Deep Learning”, Springer.

Reference books:
1. Nikhil Buduma, Nicholas Locascio- “Fundamentals of Deep Learning”, O’Reilly, Shroff
Publishers.
2. Ian J. Goodfellow, Yoshua Bengio, Aaron Courville, “Deep Learning”, MIT Press, 2017.

Course objectives and Course outcomes mapping:


● To introduce basic concepts of neural network and to comprehend shallow learning
and deep learning techniques: CO1, CO2
● To understand the effect of various hyperparameters on model’s performance: CO3
● To amplify Convolution Neural Network from the scratch and use of transfer learning
for various applications: CO4, CO5
● To learn the Recurrent Neural Network and its usage for various applications: CO6

Course units and Course outcomes mapping:


Unit Course Outcomes
Unit Name
No. CO1 CO2 CO3 CO4 CO5 CO6
1 Introduction to Neural Networks ✓
2 Shallow Neural network and Deep ✓
Neural Networks
3 Deep Neural Networks -
Hyperparameter Tuning, Regularization ✓
and Optimization
4 Convolutional Neural Network ✓
5 Sequence Models - Recurrent Neural

Networks
6 Advance Topics ✓

Programme outcomes:
PO 1: Engineering knowledge: An ability to apply knowledge of mathematics,
science, and engineering.
PO 2: Problem analysis: An ability to identify, formulates, and solves engineering
problems.
PO 3: Design/development of solutions: An ability to design a system, component,
or process to meet desired needs within realistic constraints.
PO 4: Conduct investigations of complex problems: An ability to use the techniques,
skills, and modern engineering tools necessary for solving engineering
problems.
PO 5: Modern tool usage: The broad education and understanding of new
engineering techniques necessary to solve engineering problems.
PO 6: The engineer and society: Achieve professional success with an understanding
and appreciation of ethical behaviour, social responsibility, and diversity, both
as individuals and in team environments.
PO 7: Environment and sustainability: Articulate a comprehensive world view that
integrates diverse approaches to sustainability.
PO 8: Ethics: Identify and demonstrate knowledge of ethical values in non-
classroom activities, such as service learning, internships, and field work.
PO 9: Individual and team work: An ability to function effectively as an individual,
and as a member or leader in diverse teams, and in multidisciplinary settings.
PO 10: Communication: Communicate effectively on complex engineering activities
with the engineering community and with society at large, such as, being able
to comprehend and write effective reports and design documentation, make
effective presentations, and give/receive clear instructions.
PO 11: Project management and finance: An ability to demonstrate knowledge and
understanding of the engineering and management principles and apply these
to one’s own work, as a member and leader in a team, to manage projects and
in multidisciplinary environments.
PO 12: Life-long learning: A recognition of the need for and an ability to engage in life-
long learning.
Programme outcomes and Course outcomes mapping:
Programme Course Outcomes
Outcomes CO1 CO2 CO3 CO4 CO5 CO6
PO1 ✓
PO2 ✓
PO3 ✓
PO4 ✓
PO5 ✓
PO6 ✓
PO7 ✓
PO8 ✓
PO9 ✓
PO10 ✓
PO11 ✓
PO12 ✓

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy