STAT598N • Statistics
| Course Requirements | Schedule of Readings
In order for you to participate in the class
it is essential that you've actually read the papers. While you don't
necessarily have to understand all of them, you should have at least
attempted to get through the paper,
at least looking at all of it.
To help motivate the class discussion,
everyone enrolled in the class is required to submit a short (1/2 page,
no more than 1 page; single-spaced, 12pt font) evaluation of the
being discussed. These evaluations should briefly summarize the main
contributions of each paper, as well as your assessment of its main
strengths and weaknesses. In particular, you should highlight what you
believe to be the novel insight or approach, and how it might be useful
outside the scope of the paper. These evaluations should be submitted
via email the evening before class. Since I realize many of you may
prefer to work while others of us sleep, the "evening before class"
will be construed to end at 6am, EST, on the morning of class. Those
students presenting the paper are excused from submitting an evaluation.
You may elect to skip three classes'
of reviews, in addition to presentation days, without penalty.
The most important requirement is active
participation in class discussions. The evaluations should provide
ample topics for discussion, but don't feel constrained to limit your
comments to those expressed in your evaluations. Questions or
clarifications about confusing parts of the papers are entirely
criticism or extension of
the work presented in the paper is highly encouraged. We've all
the papers---it's your new insight we're interested in!
Everyone enrolled in the class will be
asked to present at least one paper during the quarter (the more folks
enrolled, the fewer papers you'll need to present, so encourage your
friends to sign up!). This doesn't need to be a conference-quality
talk, so don't worry too much about it. The basic idea is to present to
the class a brief summary of the paper (what problem does it attempt to
tackle, why is that problem interesting, what is the approach, how
effective was it, etc.) for the benefit of those sitting in (there will
usually be several), and to present your evaluation of the strengths
and weaknesses of the paper, as well as any interesting next steps or
related issues. Think of it as an oral presentation of what you would
have written down. You should also prepare a short list of discussion
topics to help get the class started. Slides are encouraged but not
required. In many cases, it might be possible to google for slides
either on the webpages of the first authors or elsewhere.
A list of potential topics will be posted
to here or discussed in class, but feel free to suggest your own. Some
projects might be preliminary investigations of something that
could turn into conference submissions. That level of
originality is encouraged but not required. I'd be happy to
discuss such prospects with you during or after the term. You may work
on this project individually or in small groups (no more than ). A
project is expected to be larger in scope than an individual's project,
and will be graded on that assumption.
Preliminary report: Due Mar 10
The preliminary report should be a 3-4 page
report that includes:
- An overview of the project's goals and/or hypotheses,
- A description of the algorithms or models that you will
- A description of your implementation and initial data exploration
- A brief review of related work.
Presentation: Due Apr 28-30
The oral presentation to the class should provide
an overview of the
project's goals, necessary background information (including work done
by others), a description of any experiments, and a summary of your
results and findings.
Final report: Due May 5
The final project report should be a 8-15 page,
single column report. Please use 12-point font with
standard margins. Here are some useful links:
"How to write a good research
paper", Simon Peyton Jones, [pdf]
"The Science of Scientific Writing",
George D. Gopen and Judith A. Swan, American Scientist
78(6) 550-558, November-December 1990. [pdf]
Local resources: The Writing Lab at
Each paper evaluation must be completed
individually; you must write
everything that you submit. You may (and, in fact, are encouraged) to
discuss the papers with others, but you may not copy evaluations from
someone else or make your writeups available to others.
You are expected to be aware of the Department's
guidelines. Any violation of the course or institute policies will be
treated very seriously, and could lead to severe repercussions, up to,
and including, expulsion. Don't cheat. It's not worth it. For a quick
summary of department and university policies, please refer to Prof.
Spafford's webpage here.