|
|
| YZM4038 | Deep Learning | 2+0+0 | ECTS:4 | | Year / Semester | Spring Semester | | Level of Course | First Cycle | | Status | Elective | | Department | DEPARTMENT of SOFTWARE ENGINEERING | | Prerequisites and co-requisites | None | | Mode of Delivery | | | Contact Hours | 14 weeks - 2 hours of lectures per week | | Lecturer | Dr. Öğr. Üyesi Mustafa Hakan BOZKURT | | Co-Lecturer | | | Language of instruction | Turkish | | Professional practise ( internship ) | None | | | | The aim of the course: | | This course content includes understanding the concept of deep learning, learning how different deep learning models work, applying deep learning configurations, deep learning data preparation and gaining experience working with different deep neural network models. |
| Learning Outcomes | CTPO | TOA | | Upon successful completion of the course, the students will be able to : | | | | LO - 1 : | Gain programming skills for deep learning with Python language | 4,8 | 1, | | LO - 2 : | Understand classical shallow neural networks, deep neural networks and their differences | 4,8 | 1, | | LO - 3 : | Understand the differences and uses of deep neural network models | 4,8 | 1, | | LO - 4 : | Gain experience working with different data structures | 4,8 | 1,6, | | LO - 5 : | Evaluate the results obtained after neural network training | 4,8 | 1, | | CTPO : Contribution to programme outcomes, TOA :Type of assessment (1: written exam, 2: Oral exam, 3: Homework assignment, 4: Laboratory exercise/exam, 5: Seminar / presentation, 6: Term paper), LO : Learning Outcome | | |
| Introduction to deep learning. Learning process in neural networks. Single layer networks, multilayer networks. Software technologies used in deep learning. Convolutional neural networks and application. Recurrent neural networks and their applications. Long-short term memory networks and applications. Generative networks and applications. Examination and evaluation of applications of different problem types with deep neural networks. |
| |
| Course Syllabus | | Week | Subject | Related Notes / Files | | Week 1 | Introduction to Deep Learning | | | Week 2 | Learning in Neural Networks | | | Week 3 | Fully Connected Neural Networks | | | Week 4 | Fully Connected Neural Networks | | | Week 5 | Deep Learning Libraries | | | Week 6 | Convolutional Neural Networks | | | Week 7 | Classification with Convolutional Neural Network | | | Week 8 | Network Optimization and Configurations | | | Week 9 | Midterm Exam | | | Week 10 | Recurrent Neural Networks | | | Week 11 | Long-Short Term Memory | | | Week 12 | Forecasting with Long Short-Term Memory | | | Week 13 | Generative Adversarial Networks | | | Week 14 | Generative Adversarial Networks | | | Week 15 | Neural network training and interpretation of results | | | Week 16 | Final Exam | | | |
| 1 | Derin Öğrenme, Ian Goodfellow, Yoshua Bengio, Aaron Courville (2018) Buzdağı Yayınevi | | | 2 | Python ile Derin Öğrenme, François Chollet (2021), Buzdağı Yayınevi | | | 3 | Scikit-Learn, Keras ve TensorFlow ile Uygulamalı Makine Öğrenmesi, (2021), Aure?lien Ge?ron
Buzdağı Yayınevi | | | |
| Method of Assessment | | Type of assessment | Week No | Date | Duration (hours) | Weight (%) | | Project | 15 | | 1 | 40 | | End-of-term exam | 16 | | 3 | 60 | | |
| Student Work Load and its Distribution | | Type of work | Duration (hours pw) | No of weeks / Number of activity | Hours in total per term | | Yüz yüze eğitim | 2 | 14 | 28 | | Sınıf dışı çalışma | 2 | 14 | 28 | | Proje | 2 | 14 | 28 | | Dönem sonu sınavı için hazırlık | 1 | 14 | 14 | | Dönem sonu sınavı | 3 | 1 | 3 | | Total work load | | | 101 |
|