Each student is required to present one paper over the course of the semester. Since the number of students are much larger than the number of papers, there are basically three different roles that you can select:
- Presenters (2 students). Each presentation should be clear, well organized and very technical, and roughly 20 minutes long. The presenters should read the assigned paper in detail and be prepared to effectively lead the class discussion on the paper. To prepare your presentation, you can use any presentation tool (e.g., Powerpoint, Keynote, LaTex) provided that the tool has options to export the slides to PDF. You are allowed to reuse the material already exist on the web as long as you clearly cite the source of the media that you have used in your presentation.
Deadline: You should meet with the instructor 3-4 days before the presentation date to discuss your slides, and the presentation should be submitted by the night before the class.
- High-level overview of the paper (main contributions)
- Problem statement and motivation (clear definition of the problem, why it is interesting and important)
- Key technical ideas (overview of the approach)
- Experimental set-up (datasets, evaluation metrics, applications)
- Strengths and weaknesses (discussion of the results obtained)
- Connections with other work (how it relates to other approaches, its similarities and differences)
- Future direction (open research questions)
The presentations will be graded according to this rubric.
- Code walker (1 student). This student will walk through the implementation of the paper. They should connect the dots between the implementation details given in the paper and the actual code, and moreover discuss these implementation choices by relating them to the topics covered in the class. The walkthrough should be roughly 8 mins long and you should try your best to find workarounds or fixes which are not discussed in the paper.
- Demonstrator(s) (1 or 2 students) These student will conduct experiments demonstrating how the method works in practice, for which they will have about 8 mins to both present and discuss their results. And more importantly, they should try to find some hard example inputs to fool the network models.
In the Spring 2018 semester, we will read and discuss these papers.