Class time: Tuesdays, 8:40-11:30 am
Class location: EF Building
Office hours: Tuesdays, 4:00-5:00 pm
Syllabus: please check SIS.
COURSE GOALS:
The course is introductory level deep learning course, suitable for senior level undergraduate students as well as the graduate students. It will cover the topics of deep learning. Tentative topics include:
Neural Networks
Convolutional Neural Networks
RNN, LSTM-based Deep Networks for sequences in vision
Deep Reinforcement Learning
Optimization and hyperparameter tuning for vision
Encoder-decoder networks, Siamese networks, LeNet, U-Net, etc.
Generative adversarial networks
RNNs, LSTMs, GRUs,
Reinforcement learning
PRE-REQUEST:
For the official prerequests, students should check the SIS. In general, all the students should have basic understanding of probabilitiy, linear algebra and vector calculus and good working knowledge of programming (python is needed in this course).
GRADING:For most accurate grading scheme, students must refer to the syllabus. Programming assignments: total 10% of the final grade.
In programming assignments, it is expected that the submitted code is running without any error and generating the correct result(s) as described in the assignments. Codes giving error, will not be graded.
Final project: 40% (there is a 5% for the missing progress report. The progress report format will be available on LMS). Projects will be presented at the end of semester.
Mid-Term Exam 25%. (tentative date: First week of October 2018, in-class, written)
Paper presentation: 25% of the final grade.
See the guidelines on LMS for both the final project and paper presentations.
RECOMMENDED BOOKS (optional)
Ian Goodfellow, Yoshua Bengio and Aaron Courville, Deep Learning, MIT Press, 2016.
Strang, Gilbert. Linear Algebra and Its Applications 2/e, Academic Press, 1980.
PROGRAMMING
Python will be main programming environment for the assignments. Following book (Python programming samples for computer vision tasks) is freely available and is one of the good starting points with computer vision applications. Python for Computer Vision
COLLABORATION POLICY
Collaboration on assignments is encouraged at the level of sharing ideas and technical conversation only. Please write your own code. Students are expected to abide by OzU's academic integrity rules.
LECTURE NOTES
Lecture notes are updated weekly on LMS. Please check LMS for the updated lecture notes.
PAPER PRESENTATION
After the midterm, you will present your chosen paper in the class-room. This will be a group presentation relevant to your final project (you should select a paper relevant to your project).
Please submit your paper presentation details as text first. Each project group will present one paper.
In your text submission, you need to include:
Paper's info:
Full title of the paper as shown on the paper,
Complete author list of the paper,
and where the paper is published (peer-reviewed) along with its publication year,
Paper's abstract,
Paper's link to download,
Also include your own project title and your project's one sentence description in another paragraph.
The paper presentation will take 25% of your final grade. Please choose a paper relevant to your project title. That 25% distributes as follows:
Paper proposal submission: 2% (if your proposed paper title is not approved by the given deadline, you will lose this 2% as penalty),
Presentation: 23%, (if you do not submit any paper and if you do not get approval of the instructor before the presentation date, you cannot get any of this credit)
Presenting "the content" properly: 15%,
Presentation quality & skills: 8%,
Presentation details: For "the content" of the paper that you picked, you will focus on the following details and will present them in your presentation:
Check for the code of the paper on GitHub or on another online platform and download and run the code to generate author's results. (5%)
Understand the goal of the paper: What is new, and what was done previously (as described in the paper). (2%)
What are the details of the proposed solution and the architecture presented in the paper, (3%)
What metrics used in the paper and their definitions, (3%)
What are the results, and how they are presented. (2%)
BONUS: apply the code on another large data-set that was not used in the paper and generate results for that new datasets by using the performance metrics of your selected paper. (+%3 on your final grade)
Expect to have a short presentation focusing on the above-listed points (duration of your presentation will be announced in the class).