COURSE UNIT TITLE

: DEEP LEARNING METHODS AND APPLICATIONS

DEGREE PROGRAMMES

Description of Individual Course Units

Course Unit Code Course Unit Title Type Of Course D U L ECTS
DSM 5014 DEEP LEARNING METHODS AND APPLICATIONS ELECTIVE 3 0 0 8

Offered By

Graduate School of Natural and Applied Sciences

Level of Course Unit

Second Cycle Programmes (Master's Degree)

Course Coordinator

PROFESSOR DOCTOR BURCU HÜDAVERDI AKTAŞ

Offered to

Data Science
Data Science (Non-Thesis-Evening)

Course Objective

It aims to enable students to understand and use advanced artificial neural networks and deep learning methods, mainly for data science in an applied manner. This course provides an introduction to Deep Neural Networks (Deep Learning). Focusing on both theory and practice, it will cover models for various applications, how to train and test them, and how to use them in real-world applications.

Learning Outcomes of the Course Unit

1   Understand general machine learning terminology
2   Understand the motivation and operation of the most common deep neural networks
3   Describe and use various methods, algorithms, and techniques for deep learning
4   Develop various applications using deep learning application platforms such as TensorFlow, Keras, etc. that support the Python/R programming language.
5   Critically evaluate model performance and interpret results

Mode of Delivery

Face -to- Face

Prerequisites and Co-requisites

None

Recomended Optional Programme Components

None

Course Contents

Week Subject Description
1 Introduction to machine learning and neural networks: supervised learning, linear models for regression, basic neural network structure, simple examples and motivation for deep networks.
2 Introduction to Neural Networks. Biological neuron, Artificial neuron models, Perceptron learning, Multi-layer perceptron learning, Backpropagation. Supervised, semi-supervised and unsupervised learning.
3 Basics of Artificial Neural Networks: forward propagation, cost functions, error backpropagation, training by gradient descent, bias/variance
4 Basics of Artificial Neural Networks: Gradient descent, Stochastic gradient descent, other optimization methods, Issues of convergence, over fitting, regularization,generalization.
5 Applications on Perceptron and ANN
6 Introduction to Deep Learning. Auto-encoders. Learning representations with auto-encoders, influence of sparsity and issues of sparse data.
7 Convolutional Neural Networks: Basic theoretical concepts. Different layers of processing, convolution, pooling, drop-out, loss, training.
8 Applications on Convolutional Neural Networks
9 Project presentations
10 Generative Models: Segmentation, U-Net, Generative Adversarial Networks (GAN)
11 Applications on GAN
12 Deep Recurrent Neural Networks
13 Applications of Deep Recurrent Networks.
14 Project presentations. General discussion and review of the topics covered throughout the term.

Recomended or Required Reading

Y. Bengio, I. Goodfellow and A. Courville, Deep Learning, MIT Press, 2016.

L. Deng and D. Yu, Deep Learning: Methods and Applications , Foundations and Trends in Signal Processing, Now Publishers, 2014.

Supplementary Book(s): C. Bishop, Pattern Recognition and Machine Learning, Springer, 2006.

K. P. Murphy, Machine learning: a probabilistic perspective, MIT press, 2012.

Planned Learning Activities and Teaching Methods

The course is taught in a lecture, class presentation, applied examples and exercises by using tools, and discussion format. Besides the taught lecture, group presentations are to be prepared by the groups assigned and presented in a discussion session. In some weeks of the course, results of the homework given previously are discussed.

Assessment Methods

SORTING NUMBER SHORT CODE LONG CODE FORMULA
1 ASG ASSIGNMENT
2 MTE MIDTERM EXAM
3 PRJ PROJECT
4 FCG FINAL COURSE GRADE ASG * 0.30 + MTE * 0.40 + PRJ * 0.30


Further Notes About Assessment Methods

None

Assessment Criteria

To be announced.

Language of Instruction

Turkish

Course Policies and Rules

Basics of probability and statistics theory, linear algebra, calculus and programming skills are needed.

Contact Details for the Lecturer(s)

Burcu Hudaverdi
Faculty of Scieinces, Department of Statistics

burcu.hudaverdi@deu.edu.tr

Office Hours

Will be announced.

Work Placement(s)

None

Workload Calculation

Activities Number Time (hours) Total Work Load (hours)
Lectures 14 3 42
Preparing assignments 2 30 60
Preparing presentations 2 12 24
Preparations before/after weekly lectures 13 4 52
Preparation for final exam 1 20 20
Project Final Presentation 1 2 2
Midterm 1 2 2
TOTAL WORKLOAD (hours) 202

Contribution of Learning Outcomes to Programme Outcomes

PO/LOPO.1PO.2PO.3PO.4PO.5PO.6PO.7
LO.13534445
LO.25345445
LO.32454535
LO.44334343
LO.55434444

CONTACT INFORMATION
Dokuz Eylül Üniversitesi Cumhuriyet Bulvarı No: 144 35210 Alsancak / İZMİR
Phone: +90(232) 412 12 12 - Fax: +90 (232) 464 81 35

Copyright 2015 DEÜ. BİD

HOME / ABOUT US / DEGREE PROGRAMMES / GENERAL INFORMATION FOR STUDENTS / INTERNATIONAL RELATIONS