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

PES UNIVERSITY

(Established under Karnataka Act No. 16 of 2023)


100 ft. Ring Road, Bengaluru – 560085, Karnataka, India

Special Topic

Course: Deep Learning and Its Applications


PES UNIVERSITY Department of Computer Applications

Course Overview
The course Deep Learning and Its Applications provides a comprehensive intro-
duction to Deep Learning, a powerful and rapidly evolving field within Artificial
Intelligence (AI). Deep learning techniques are revolutionising numerous fields,
from computer vision and natural language processing to healthcare and anomaly
detection. This course equips the students with the fundamental knowledge and
hands-on experience to understand how deep learning works, explore various ar-
chitectures like CNNs and RNNs, and implement their own deep learning appli-
cations. Through this course, the students also gain valuable skills not only to
understand the current landscape of deep learning but also to prepare them for
future advancements in this transforming technology.

Page 1 of 5
PES UNIVERSITY Department of Computer Applications

Course Objectives

Course Objectives
The objectives of the course are to help students:

1. Gain fundamental understanding of deep learning concepts, including ANNs,


activation functions, and their applications in various machine learning tasks

2. Explore the architectures and functionalities of CNNs and RNNs and un-
derstand the uses of transfer learning

3. Explore various deep learning hyperparameters, tuning techniques, and train-


ing strategies

4. Gain understanding of applications of deep learning in various domains and


emerging areas of deep learning

Course Outcomes
Upon completion of the course, the students shall be able to:
1. Explain the fundamental concepts of Artificial Neural Networks (ANNs),
including activation functions, weights, biases, and their roles in deep learn-
ing

2. Explain and use CNNs and RNNs for various tasks and apply transfer learn-
ing for deep learning tasks

3. Analyse and compare different optimisation algorithms, and hyperparam-


eter settings for training deep learning models, and apply various training
strategies for deep learning

4. Apply deep learning techniques in various domains, considering ethical


concerns associated with such applications

Page 2 of 5
PES UNIVERSITY Department of Computer Applications

COURSE OUTLINE AND CONTENT

Unit Title Topics


((L + T) hrs.)
1 Essentials of Deep Learning Types of ML and Application of
(4 + 3 hrs.) Deep Learning in Different Types
of ML
Introduction to ANNs, Activation
Functions, Weights and Biases,
Different Areas of Deep Learning
Introduction to Deep Learning
Frameworks
2 Convolutional Neural Networks Building blocks of CNNs: Filters,
and Recurrent Neural Networks Kernels, Normalisation, Pooling,
(4 + 3 hrs.) Dimensionality Reduction
Variants of RNNs: Long-Short
Term Memory (LSTM), Gated Re-
current Unit (GRU)
Comparison of CNNs and RNNs,
Transfer Learning
3 Training Strategies and Hyperpa- Exploring different optimisation
rameter Optimisation algorithms: Adam, Nadam, RM-
(4 + 3 hrs.) SProp, SGD
Other Hyperparameters: Dropout,
Regularisation, Learning Rate
Training Strategies: Holdout, k-
fold Stratified Cross-Validation,
Iterative Holdout
4 Applications of Deep Learning Introduction to Generative Adver-
(4 + 3 hrs.) sarial Networks (GANs) and Nat-
ural Language Processing (NLP)
Domain Applications: Deep
Learning in Healthcare, Deep
Learning in Anamoly Detection

Page 3 of 5
PES UNIVERSITY Department of Computer Applications

EXPERIENTIAL LEARNING

Each unit shall have an experiential learning component which enables the
students to gain hands-on experience of deep learning and its applications.

Unit Title Experiential Learning


1 Essentials of Deep Learning Introduction to Deep Learning
Frameworks: Keras and Tensor-
Flow
Classes and Functions of Keras
and TensorFlow
2 Convolutional Neural Networks Building CNN and RNN models
and Recurrent Neural Networks using TensorFlow for various deep
learning tasks
Using state-of-the-art models with
transfer learning for deep learning
tasks
3 Training Strategies and Hyperpa- Training models with different hy-
rameter Optimisation perparameters and various training
strategies
4 Applications of Deep Learning Case Study: Deep Learning in
Healthcare

Page 4 of 5
PES UNIVERSITY Department of Computer Applications

TEXT BOOKS AND REFERENCES

Textbooks:

1. Deep Learning Illustrated, John Krohn, Grant Beyleveld, and Agale´ Bassens,
O’Reilly, 2019. ISBN: 9780135116821

Reference Books:

1. Deep Learning for Dummies, John Paul Mueller and Luca Massaron, Wiley,
2019. ISBN: 978-1-119-54304-6

Documentation:

1. Keras 3 API Documentation: https://keras.io/api/

2. TensorFlow Essential Documentation: https://www.tensorflow.


org/guide

Page 5 of 5

You might also like