Announcements

  • No announcements

General

Lecturers: Dr. Michael T.M. Emmerich & Dr. Andre Deutz
Lectures: Wednesday 11:15-13:00, Snellius WI-405.
Exam date:

Course description

Optimization algorithms are used to find parameters (or structures) that maximize or minimize user-defined objective functions. Applications of optimization algorithms can be found in machine learning and systems design. For some optimization problems, efficient algorithms are known, while for many non-linear or discrete problems heuristics are used to solve them. The development of such heuristics is an area of ongoing research.

In practice it is often the case that there are more than one objectives. Solving problems with multiple, conflicting objectives requires adaptations of classical optimization algorithms that can work in partially ordered problem space. In contrast to single objective optimization, a set of solutions is computed that can be interpreted as a trade-off curve or surface.

This class provides an introduction in the working principles of classical and modern single- and multioobjective optimization algorithms. The application of these techniques will be demonstrated for examples in bio- and chemoinformatics, such as molecular docking/folding, molecular design, inverse problems for nonlinear systems, and error minimization in machine learning.

Topics

  1. Overview on Optimization Problems (Lecture)
  2. One-dimensional Optimization Algorithms (Lecture)
  3. N-dimensional Optimization Algorithms (Lecture)
  4. Heuristic Optimization (Lecture)
  5. Multiobjective (Pareto) Optimization (Lecture)
  6. Applications in Bioinformatics (Student presentations)

Course regulations

Grade = 0.2 * Presentation + 0.2 * Practical Assignment + 0.6 * Exam

Course material

Lecture Slides

  1. Optimization
  2. Optimization under constraints
  3. Searchspace
  4. 1D Algorithms and methods
  5. nD Algorithms and methods
  6. Gradient and Newton methods

Recommended literature

Schedule

DateTimeRoomLectureTopic
Feb. 411:15-13:00405LectureIntroduction
Feb. 1111:15-13:00405Lecture--
Feb. 1811:15-13:00405Lecture--
Feb. 2511:15-13:00405Lecture--
Mar. 411:15-13:00405Lecture--
Mar. 1111:15-13:00405Lecture--
Mar. 1811:15-13:00405Lecture--
Mar. 25--------
Apr. 111:15-13:00405Lecture--
Apr. 811:15-13:00405Lecture--
Apr. 1511:15-13:00405Lecture--
Apr. 2211:15-13:00405Seminar--
Apr. 2911:15-13:00405Seminar--
May. 6--------
May. 1311:15-13:00405Seminar--