EGE UNIVERSITY

GRADUATE SCHOOL OF NATURAL AND APPLIED SCIENCES
ENGINEERING SCIENCES

COMPUTER ENGINEERING DEPARTMENT

 

 

 

 

2019-2020 SPRING SEMESTER

 

 

 

 

Course

 

 

618 DEEP LEARNING (3+0)

 

 

Instructor

 

 

Prof. Dr. Aybars UĞUR

 

Course Place and Time

 

 

EGE University Computer Engineering Department B3 Room
(
Tuesday, 13:15-16:00)

 

 

Learning Outcomes

 

 

 

1

To be able to design and implement Deep Learning system.

2

To do research in state-of-the-art subjects of Deep Learning area; preparing and doing presentation. To gain experience in reading and writing papers in Deep Learning.

3

To learn ability to use and integrate Software Tools in Artificial Neural Networks, Machine Learning and Deep Learning.

4

To learn basic concepts and techniques of Deep Learning, mathematical and software background; to have ability to apply Deep Learning to problems. To have both a general “breadthknowledge of Deep Learning and Machine Learning techniques, plus a deeper specialized knowledge of one particular sub-area; how to combine or integrate them.

 

Description

 

Introduction to Deep Learning, Machine Learning Paradigms, Artificial Neural Networks, Ensemble Learning Methods, Convolutional Deep Neural Networks (CNN), Recurrent Neural Networks (RNN), Long Short Term Memory (LSTM), Deep Autoencoders, Other Deep Learning Methods, Hybrid Intelligent Systems.

 

 

Syllabus

 

 

Week #

1

Introduction to Deep Learning, concepts, terminology and methods.

2

Machine Learning - I : Supervised Learning, Unsupervised Learning, Reinforcement Learning, Semi-Supervised Learning; Supervised Learning Methods (Neural Networks and Backpropagation Algorithm, Decision Trees, Naive Bayes, Support Vector Machines, ...); Ensemble Learning Types (Bagging, Boosting, Stacking) and Methods (Random Forests, AdaBoost, ...)

3

Machine Learning - II : Unsupervised Learning and Clustering Methods, Reinforcement Learning Methods.

4

Deep Learning – I : Convolutional Neural Networks (CNN) (Architecture, Datasets, Convolution, Mechanisms), Application development in Tensorflow environment.

5

Deep Learning – II : Recurrent Neural Networks (RNN), Long Short Term Memory (LSTM).

6

Deep Learning – III : Deep Belief Networks, Deep Autoencoders, Other Deep Learning Models.

7

Deep Learning - IV : Application Areas of Deep Learning, Hybrid Intelligent Systems.

8

Midterm Exam ? #

9

Presentations - I

10

Presentations - I

11

Presentations - I

12

Presentations - II

13

Presentations - II

14

Presentations - II

15

Project Demos

16

Final Exam

 

 

 

Textbook

 

 

·         Goodfellow, Y. Bengio and A. Courville, “Deep Learning”, MIT Press, 2016.

 

·         Ian Goodfellow, Yoshua Bengio, Aaron Courville, “Derin Öğrenme”, Buzdağı Yayınları, 2018 (In Turkish) 

 

 

Reference Books

 

·         Deep Learning with Python, François Chollet, 1st edition, Manning Publications.

 

·         Deniz KILINÇ, Nezahat BAŞEĞMEZ, Uygulamalarla Veri Bilimi, Abaküs Yayın, 2018.

 

·         Michael Negnevitsky, “Artificial Intelligence : A Guide to Intelligent Systems (3rd Edition)”, Addison Wesley, 2011.

 

Also Sample Papers will be given.

 

 

Links

 

https://youtu.be/njKP3FqW3Sk

 

 

Prerequisites

 

Artificial Intelligence and Machine Learning Basics:  Problem solving, state space search, machine learning principles, pattern recognition, fundamentals of computer vision. Basic knowledge of probability, statistics, calculus and linear algebra.

 

Only for PhD Students.

 

 

Grading

 

Midterm Activities (50%)

·         Project (+ Paper) (30%)

·         Presentation 1 (10%)

·         Presentation 2 (10%)

·         Take-Home -

 

Final Exam (50%)

 

For the Take-Home and reports, students will be allowed a total of 5 (five) late days; each additional late day will incur a 10% penalty.