ECTS @ IUE ECTS @ IUE ECTS @ IUE ECTS @ IUE ECTS @ IUE ECTS @ IUE ECTS @ IUE

Syllabus ( CSE 655 )

   Basic information
Course title: Deep Learning and Applications
Course code: CSE 655
Lecturer: Assoc. Prof. Dr. Yakup GENÇ
ECTS credits: 7.5
GTU credits: 3 (3+0+0)
Year, Semester: 1, 2, Fall and Spring
Level of course: Second Cycle (Master's)
Type of course: Area Elective
Language of instruction: English
Mode of delivery: Face to face
Pre- and co-requisites: none
Professional practice: No
Purpose of the course: Deep learning, a branch of machine learning, allows computers to model high-level abstractions from experience (encoded in large-scale labeled and unlabeled data). Recent advances in computing hardware and algorithms have made it a popular tool for artificial intelligence. This course aims at clarifying the theory behind deep learning methods while providing the students with the skills of their effective use in many domains such as computer vision and natural language processing.
   Learning outcomes Up

Upon successful completion of this course, students will be able to:

  1. Basic knowledge of machine learning and deep learning methods.

    Contribution to Program Outcomes

    1. Define and manipulate advanced concepts of Computer Engineering
    2. Formulate and solve advanced engineering problems
    3. Review the literature critically pertaining to his/her research projects, and connect the earlier literature to his/her own results
    4. Work effectively in multi-disciplinary research teams
    5. Find out new methods to improve his/her knowledge.
    6. Effectively express his/her research ideas and findings both orally and in writing

    Method of assessment

    1. Written exam
    2. Homework assignment
    3. Seminar/presentation
    4. Term paper
  2. Knowledge and experience on how to apply deep learning methods in various domains such as computer vision, natural language processing and big data.

    Contribution to Program Outcomes

    1. Define and manipulate advanced concepts of Computer Engineering
    2. Review the literature critically pertaining to his/her research projects, and connect the earlier literature to his/her own results
    3. Work effectively in multi-disciplinary research teams
    4. Find out new methods to improve his/her knowledge.
    5. Defend research outcomes at seminars and conferences.

    Method of assessment

    1. Written exam
    2. Homework assignment
    3. Seminar/presentation
    4. Term paper
  3. Knowledge of literature with a focus on recent developments in deeep learning.

    Contribution to Program Outcomes

    1. Define and manipulate advanced concepts of Computer Engineering
    2. Work effectively in multi-disciplinary research teams
    3. Continuously develop their knowledge and skills in order to adapt to a rapidly developing technological environment,
    4. Effectively express his/her research ideas and findings both orally and in writing
    5. Develop awareness for new professional applications and ability to interpret them

    Method of assessment

    1. Written exam
    2. Homework assignment
    3. Seminar/presentation
    4. Term paper
   Contents Up
Week 1: Intro to Machine Learning
Week 2: Machine Learning Basics
Week 3: Deep Learning Tools - Caffe, Torch, TensorFlow, Theano
Week 4: Feedforward Deep Networks
Week 5: Regularization of Deep or Distributed Models
Week 6: Optimization for Training Deep Models
Week 7: Convolutional Networks
Week 8: Sequence Modeling: Recurrent and Recursive Nets
Week 9: Structured Probabilistic Models for Deep Learning
Week 10: Linear Factor Models and Auto-Encoders
Week 11: Application in Computer Vision
Week 12: Application in Big Data
Week 13: Application in Natural Language Processing
Week 14: Application in Speech Processing
Week 15*: -
Week 16*: Final Exams
Textbooks and materials: Deep Learning by Yoshua Bengio et al MIT Press, 2015.
Recommended readings: Hands on Machine Learning with Scikit-Learn & Tensorflow by Aurelien Geron, 2019.
  * Between 15th and 16th weeks is there a free week for students to prepare for final exam.
Assessment Up
Method of assessment Week number Weight (%)
Mid-terms: 0
Other in-term studies: 0
Project: 4 40
Homework: 1 30
Quiz: 0
Final exam: 16 30
  Total weight:
(%)
   Workload Up
Activity Duration (Hours per week) Total number of weeks Total hours in term
Courses (Face-to-face teaching): 3 14
Own studies outside class: 5 14
Practice, Recitation: 0 0
Homework: 6 5
Term project: 5 7
Term project presentation: 5 1
Quiz: 0 0
Own study for mid-term exam: 0 0
Mid-term: 0 0
Personal studies for final exam: 0 0
Final exam: 2 1
    Total workload:
    Total ECTS credits:
*
  * ECTS credit is calculated by dividing total workload by 25.
(1 ECTS = 25 work hours)
-->