Information for UVA Qdata Group's Deep2Read Page
About this website:
- As a group, we need to improve our knowledge of the fast-growing field of deep learning
- To educate new members with basic tutorials, and to help existing members understand advanced topics.
- This website includes a (growing) list of tutorials and papers we survey for such a purpose.
- We hope this website is helpful to people who share similar research interests or are interested with learning advanced topics about deep learning.
- Please feel free to email me (email@example.com), if you have related comments, questions or recommendations.
- BTW: The covered tutorials and papers are by no means an exhaustive list, but are topics which we have learned or plan to learn in our reading group.
Course Basics: and General Description
- This is an advanced graduate-level deep learning course.
- The course takes the form of half-seminar and half-project. The form of seminar focuses on paper readings.
- This course offers opportunities for students to get into research topics about the
state-of-the-art advanced deep learning.
- No text book
- Sit-in: No. This course is for registered students only.
- This website was started from the seminar course at UVA I taught in Fall 2017 and Spring 2019.
- This course offers opportunities for students to have in-depth understanding and hands-on experience of deep learning. Students are expected to generate top-tier publications when finishing the course.
- Instructor’s Permission for enrollment is required for this course.
- Required courses as prerequisite: Graduate-level machine
learning; Introduction of Deep Learning and Graduate-level Optimization are preferred.
- Familiar reading of Basic Deep Learning are preferred.
Course Grading Policy
The grade will be calculated as follows:
- 60% for the in-class paper presentations/discussions/ note taking
- 40% for the project
- Sharelatex/overleaf to submit lectures about the assigned papers
Each class, we will assign 4 to 6 reading materials (video lectures or papers or research lecture slides )
Each student is expected to have three sets of assigntments:
- (a) Weekly project summary should be updated per week right before project meetings with Prof. Qi;
- (b) Assigned presentation slides: please use the BEAMER template shared through the course overleaf project. Please make sure the presentation slides are ready before every Friday 8am; (One Example Slide Presenation)
(c) Assigned scribe notes: please use the latex template shared through the course overleaf project. Please make sure the scribe notes are ready one week after. (One Example Scribe note)
- For both the paper presentations and the scribe notes, please use the following structure as reference:
- Full reference of the paper
- Motivations / Why needed ? / Why important ?
- Previous solutions
- Key insights
- Key equations
- Key conclusions
- Goals achieved: / Under what restrictions or assumptions;
- Announcements are being emailed to the course mailing list.
- A welcome note will be sent to the mailing list early in the semester.
- Errata and answers to questions are being discussed and answered
on the course emailist.