Some of the course meeting times will be reserved for discussing papers relevant to the lecture material. We will discuss two papers in one meeting time. The papers are available through the Cornell library, and electronic versions will also be posted on the course website. All registered students, formal listeners, and informal listeners are expected to carefully read assigned papers and come to class prepared to discuss each paper's strengths, weakness, and directions for future work. Reading guides will be posted on the course website to help focus the student's time on the most relevant portions of each paper. The reading guides will often include a related paper and will note whether students are expected to briefly skim or read in detail this related work. In addition, students are required to prepare several short paper critiques and one long paper critique as described below.
Mini-Paper Critique #1 Assignment | GEOG 587: Conservation GIS
Students will spend the final portion of the semester working in small groups to design, implement, and evaluate an interesting research idea related to chip-level interconnection networks. Groups should include exactly two students if at all possible. Students are strongly encouraged to select projects that overlap with their own research. For example, students working on emerging technologies might study the implication of these technologies on interconnection networks; students working on digital circuits might design a network component at the circuit-level; students working on processor core architecture might study network architectures that effectively interconnect their core design; students working at the application-level might evaluate the impact of a interconnection network in a real machine when running an interesting application. Your short paper critiques as well as the critiques of the other students should provide a great source of inspiration for project ideas. I am very open to creative project ideas. Although I am much more interested in projects that tackle interesting new ideas, it is also possible for students to pick a recent paper from a top-tier conference and validate some of the paper's results. Students are on their own for setting up an appropriate evaluation framework, although I can provide some suggestions for getting started with Simics+GEMS, booksim, or my own mtlsim framework. A short three- to four-page project proposal is due before class on April 1st (same font, spacing, and margin format as the short critiques). Each group will give an 8- to 10-minute presentation on either May 4th, and submit a six- to eight-page conference-style paper by 5:00pm on May 7th. The paper should be single spaced, double column, 10pt font, with 1in margins in a similar format as the research papers we have been discussing in class. There are absolutely no extensions for the final paper.
Mini-Paper Critique #1 Assignment
Each registered student is responsible for shepherding one paper discussion. Please email me if you would like to volunteer to shepherd a specific paper, otherwise I will have to pick which students shepherd which papers. When shepherding a paper, the student should read the paper in particular detail trying to understand every aspect of the paper. The student might also need to review some of the related work mentioned in the paper. The student then gives a 10-minute presentation before we start discussing the paper in class. This presentation should briefly summarize the main points of the paper, possibly clarifying anything that the student thought was unclear. The presentation should end with some topics for discussion. Shepherding a paper then involves helping me to lead the discussion. For example, the paper shepherd might answer another student's question about some specific part of the paper. Or if the discussion seems to be lagging, then the shepherd can propose a new topic for discussion. After class, the paper shepherd reflects on the discussion, reviews the submitted short paper critiques, and possibly investigates any issues that are still unclear. This could even involve emailing the authors of the paper for clarification. The shepherd is then responsible for writing a long paper critique (four to five pages). This critique should be similar in spirit and format as the short paper critiques but with much greater detail. Long paper critiques are due via email, no later than two weeks from the day of discussion. When shepherding a paper, the student does not need to submit a short paper critique for that day of discussion.
Paper critique as an educational method in epidemiology.