Lecture: 3 hours

Prerequisite: Course ECS 177 or 277 recommended

Grading: S/U, participation (30%), homework (30%), project (40%)

Catalog Description:
Surveys current research on topology-based visualization and data analysis techniques, including level-set methods, Morse theory based methods, and vector field methods.

Expanded Course Description:

Complex data requires sophisticated and rigorous methods for its effective analysis. Topology-based methods have recently become popular in identifying, ordering, and simplifying features in a wide range of application areas. Lectures and reading will survey current and historically relevant techniques for topology-based visualization. Students will learn several topological models and efficient algorithms that can be applied to the analysis and visualization of real valued data representing physical phenomena, such as temperature, pressure, electron density distributions and so on. An emphasis will be placed on identifying the appropriate corresponding analysis techniques for various application areas. The course material ranges from an introduction of fundamental concepts to the most cutting-edge papers in the field. Some topics that will be covered are:

  • Definition of features
  • Morse theory
  • Persistence and simplification
  • Complexes
  • Contour trees/Reeb graphs
  • Morse-Smale complex
  • Discrete Morse theory
  • Vector field methods
  • Time dependent methods

A selection of research papers will be provided along with lecture notes.

Computer Usage:
Computer usage in the projects is expected, although possibly purely theoretical projects may be acceptable.

Final Project:
A final project will be required. Projects will be devised by the students in consultation with the instructor. Final projects can take the form of (a) a research project extending current methods, with a write-up, (b) applying an existing analysis technique to a new application area, with a write-up, or (c) a survey of 4 or more related papers in a relevant area. Projects, which extend the students’ existing research are encouraged.

Each week, students will be required to read and summarize the key contributions of a research paper selected by the instructor.

Instructor: B. Hamann, A. Gyulassy

Prepared by: B. Hamann, A. Gyulassy, V. Natarajan

Overlap Statement:
This course does not have a significant overlap with any other existing course. ECS 177 also discusses on a high level, but not in depth as done here, the concept of features in scalar and vector fields; nevertheless, ECS 177 does not cover the specific methods and algorithms used in this course

Revised: 8/08


Lecture: 1 hour

Prerequisite: Graduate Standing

Grading: S/U, class participation (100%)

Catalog Description:
The primary purpose of this seminar is to keep graduate students abreast of each other’s research, as well as to bring in first-tier guest lecturers that students are unlikely to see elsewhere.

Expanded Course Description:

This weekly seminar is for the entire security lab: students, staff, and faculty. There are several purposes for the seminar:

  • A chance for everyone in the security lab to learn about the research that’s going on around them and the people who are doing the research
  • A useful and safe venue for people to give practice talks
  • A useful and safe venue for people to discuss their research and get ideas and feedback in an informal environment
  • A chance for students to hear prominent, outside guest lecturers

Many of the talks will discuss research that’s just preliminary or half-baked, but that’s encouraged. It’s often much more useful to discuss ideas that haven’t been thought out completely than when the ideas have already been codified in print. Most weeks, we hope to have one or two students present their work for between 30 minutes and 1 hour each, but we will sometimes use the time for guest presentations. Students are encouraged to volunteer to speak each week.

Textbook: None

Final Project: None

Homework: None

Instructor: M. Bishop

Prepared by: M. Bishop, S. Peisert

Overlap Statement:
This course does not have a significant overlap with any other existing course.