ETH - SOP - Education - Lecture - Bio-inspired Optimization & Design

this webpage might no longer be updated more...

Bio-Inspired Optimization & Design

Lecture Notes

Lecture notes will be sold as hard copies on February 24th.

Course number	251-0532-00L	Lecture	Tue 13-15 (CAB G 61)
Lecturers	Prof. Dr. Eckart Zitzler (Part I)	Exercises	Tue 15-16 (CAB G 61)
	Prof. Dr. Petros Koumoutsakos (Part II)	Start	17.2.2009 (Lecture)
Assistants	Johannes Bader (Project 1)		24.2.2009 (Exercises)
	Tamara Ulrich (Project 2)

Overview

Biologically-inspired computation is an umbrella term for different computational approaches that are based on principles or models of biological systems. This class of methods such as evolutionary algorithms, ant colony optimization, and swarm intelligence complements traditional techniques in the sense that the former can be applied to large-scale applications where little is known about the underlying problem and where the latter approaches encounter difficulties. Therefore, bio-inspired methods are becoming increasingly important in face of the complexity of today's demanding applications, and accordingly they have been successfully used in various fields ranging from computer engineering and mechanical engineering to chemical engineering and molecular biology. This lecture focuses on the foundations of bio-inspired computation with an emphasis on their application to optimization. The exercises will be oriented towards the implementation of these concepts to realistic applications.

Teaching Objectives

You are familiar with the foundations of optimization and with different randomized search algorithms, in particular bio-inspired ones.
You will be able to design, implement, and tune basic and advanced bio-inspired optimization techniques for tackling complex, large-scale applications.
You will be able to evaluate different search algorithms and implementations.
You are aware of the theoretical foundations of bio-inspired optimization, know the limitations as well as potential advantages and disadvantages of specific design concepts.

Testatbedingung

For getting the testat (Testatbedingung), at least 40 points need to be reached for each project and 200 points in total for all four projects. Tasks which can easily be skipped are marked as supplementary tasks.

Part I

0. Introduction and Overview

1. Optimization and Search

2. Randomized Search Algorithms

2.1 Black-Box Optimization

2.2 Local Search

2.3 Metropolis Algorithm

2.4 Simulated Annealing

2.5 Tabu Search

2.6 Evolutionary Algorithms

3. Basic Design Issues

3.1 Representation

3.2 Fitness Assignment

3.3 Selection

3.4 Variation

3.5 Example Application: Clustering

4. Advanced Design Issues

4.1 Multiobjective Optimization

4.2 Constraint Handling

4.3 Implementation Tools

4.4 Example Application: Network Processor

5. Performance Assessment

5.1 General Aspects

5.2 The No-Free-Lunch Theorem

5.3 Running Time Analysis

Part II

Part II of the lecture begins April 15th. Please refer to the website for further information.

Schedule

Date	Content	Slides	Projects
17.02.2009	0 - 2.1	Chapter 0, Chapter 1	no exercise
24.02.2009	2.2 - 2.6	Chapter 2	Project 1: Task assignment
03.03.2009	3.1 - 3.2	n/a	(Question time)
10.03.2009	3.3 - 3.5	Chapter 3	Project 1: Discussion of Task 1
17.03.2009	4.1 - 4.2	n/a	Project 1: Discussion of Task 2 Project 2: Task assignment
24.03.2009	4.3 - 5.2	Chapter 4	(Question time)
31.03.2009	5.3	Chapter 5	Project 2: Discussion of Task 1
07.04.2009	Start of Part II	n/a	Project 2: Discussion of Task 2
14.04.2009	Part II, see the websites of the ICoS group
...	Part II, see the websites of the ICoS group

Projects

		pdf	materials
Project 1	Knapsack Problem I
Project 2	Knapsack Problem II

RSS Feed

The following RSS-feed delivers all lecture notes, projects and information:

|BOD Materials