Nature Inspired Computation - 2019 entry

There are a wide range of tasks, including product design, decision making, logistics and scheduling, pattern recognition and problem solving, which traditional computation finds it either difficult or impossible to perform. However, nature has proven to be highly adept at solving problems, making it possible to take inspiration from these methods and to create computing techniques based on natural systems.  This module will provide you with the knowledge to create and apply techniques based on evolution, the intelligence of swarms of insects and flocks of animals, and the way the human brain is thought to process information.  This module is appropriate for any student with an interest in natural systems, optimisation and data analysis who has some programming and mathematical experience.


Prerequisite module: ECM1410 and ECM1414 or equivalent

This module aims to provide you with the necessary expertise to create, experiment with and analyse modern nature-inspired algorithms and techniques as applied to problems in industry and industrially motivated research fields such as operations research.


The module also aims to provide you with knowledge of the limitations and advantages of each algorithm and the expertise to determine the appropriate algorithm selection for a given problem.

On successful completion of this module, you should be able to:


Module Specific Skills and Knowledge:
1 demonstrate a clear understanding of the difficulties associated with certain intelligence-related tasks that we would wish to program computers to do;
2 describe in broad terms, the execution of each nature-inspired algorithm;
3 discuss the circumstances and environments in which each algorithm is best employed;
4 define the different underlying natural mechanisms of each algorithm and explain how this leads to improved computational performance;
5 evaluate a difficult problem and determine the likely best algorithm selection.
Discipline Specific Skills and Knowledge:
6 implement software for addressing real-world optimisation problems with nature-inspired methods;
7 create software for addressing certain complex real-world pattern recognition problems.
Personal and Key Transferable / Employment Skills and Knowledge:
8 choose appropriate techniques for given problems from a very diverse toolbox of methods;
9 explain how new ideas in science and engineering can emerge from lateral thinking and ideas from other disciplines;
10 digest and communicate succinctly information from publications in the field to individuals unfamiliar with the material.

- classical vs. nature-inspired computation;
- evolutionary algorithms (including genetic programming and multi-objective evolutionary algorithms);
- ant colony optimisation;
- particle swarm optimisation;
- swarm intelligence;
- neural computation (including multi-layer perceptrons and self-organising maps);
- artificial life;
- cellular automata;
- immune system methods.
 

Referred and deferred assessment will normally be by examination. For referrals, only the examination will count, a mark of 40% being awarded if the examination is passed. For deferrals, candidates will be awarded the higher of the deferred examination mark or the deferred examination mark combined with the original coursework mark.
 

ELE – http://vle.exeter.ac.uk

Reading list for this module: