CSCI 6514: Strategies for Search and Optimisation | |||
| Note: | This course will not be offered in
2007/2008. I teach CSCI 6604: Advanced
Computer Graphics instead. | ||
| Lectures: | TBA | ||
| Instructor: | Dirk Arnold | ||
| Office hours: | stop by whenever I'm there | ||
| Forum: | click here | ||
| Course description: | This class provides a broad overview of strategies for tackling difficult optimisation problems that occur in computer science, in the engineering sciences, and beyond. It covers "classical" approaches such as conjugate gradient strategies and dynamic programming as well as more recent, nature-inspired strategies including evolutionary algorithms and simulated annealing. | ||
| Recommended reading: |
There is no textbook for the course. The material covered is taken
from a variety of sources that will be indicated in the course of
the semester. Some of the more useful texts include:
E. Aarts and J. K. Lenstra,Local Search in Combinatorial Optimization, Princeton University Press, 2003. 
E. M. L. Beale,Introduction to Optimization, Wiley, 1988. | ||
| Grading: | Grades will be based on three mini-projects (90%) and in class participation (10%). Mini-projects consist of devising, implementing, and experimenting with optimisation strategies for given optimisation problems. As such, they involve some programming as well as a write-up of the results that have been obtained and the observations that have been made. The topics for the mini-projects will be chosen such that there is not one correct solution, but to encourage experimentation and creativity. | ||
| Projects: |
guidelines for project write-ups Project 1 Project 2 Project 3 | ||
| Slides: | 0. Introduction 1. Greedy Strategies, Dynamic Programming, Sequence Alignment, and Meta-Heuristics 2. Continuous Optimisation and Evolution Strategies 3. Strategies for Combinatorial Optimisation 4. Linear Programming, Simulated Annealing, and Neural Networks | ||