Empirical Methods


Research Methods for Empirical Computer Science

CMPSCI 691DD • Spring 2015 • Mon and Wed 10:35AM – 11:50AM • CMPS 140

DescriptionScheduleProjectReviewing ReportsResourcesSubmission System


Instructor Prof. Emery Berger
Computer Science Building 344
Office Hours TBA
Prerequisites Graduate standing in the Department of Computer Science or the Department of Electrical and Computer Engineering; At least one course in statistics; Completion of first year or some prior research project is strongly encouraged.
Description Research Methods introduces graduate students to basic ideas about conducting a personal research program. Students will learn methods for reading technical papers, selecting research topics, devising research questions, planning research, analyzing experimental results, modeling and simulating computational phenomena, and synthesizing broader theories.The course will be structured around three activities: lectures on research strategy and tactics, statistical methods, and experimental design; discussions of technical papers; and preparation and review of written assignments. Significant reading, reviewing, and writing will be required, and students will be expected to participate actively in class discussions.
Oliver, J. (1991). The Incomplete Guide to the Art of Discovery. Columbia University Press. 0-231-07620-7. (208 pages). [PDF] [PDF-alt]

Cohen, P. (1995). Empirical Methods in Artificial Intelligence. MIT Press.

Grading Criteria 40% Project reports
20% Project reviews
20% Class participation
20% Response reports

The project reports will consist of specific elements of a semester-long research project. Students will be expected to produce each report (e.g., algorithm description, literature review, experimental design). For selected project reports, students will be expected to reviewing the reports of two other students.A response report briefly summarizes the of goal of an assigned reading and then makes two or more primary points that critique, dispute, reinforce, or extend findings of the reading. See this example response report for further guidance.A single response report on the assigned readings is due by 10:00 pm of the day preceding selected classes (see the schedule). Each response report should be three paragraphs long and should be submitted using the online submission system.

Each student can opt out of the discussion in up to three classes by informing the instructor prior to the beginning of class. Grades on the lowest three response reports will be dropped before final grades are calculated.