This module introduces Intelligent Computing as a technical subject and as a field of intellectual activity. The overall targets are:- to draw a distinction between the class of problems that can be solved using traditional algorithmic approaches and those that require more sophisticated approaches that will be covered as part of this module- to present practical methods of dealing with problems that have a degree of uncertainty- to introduce problem solving approaches that have been inspired by biological evolution- to develop genetic algorithms as an alternative knowledge acquisition/representation paradigm and describe a range of genetic operators and how they can be implemented algorithmically to solve problems- to introduce optimisation techniques that are based on interaction amongst a population of embodied agents and implement solutions to simple optimisation problems using this technique- to introduce the idea of emergent behaviour or non programmed behaviour through the use of finite state machines- to introduce neural computing as an alternative knowledge acquisition/representation paradigm, to explain its basic principles and relationship to neurobiological models, to describe a range of neural computing techniques and their application areas.
Introduction
Introduction to Intelligent Computing and its importance in providing computer based solutions to real world problems. Brief description of problem solving techniques that come under the umbrella of Intelligent Computing such as Fuzzy logic, Genetic Algorithms, Particle Swarm Optimisation, Artificial Life and Neural Networks.
Fuzzy Logic
Comparison of binary logic and fuzzy logic. Description of basic fuzzy logic control system including fuzzification of input, rules, inference engine, and defuzzification of output. Case study that covers design and implementation of the heating control system based on fuzzy logic.
Genetic Algorithms
Introduction to the power of evolutionary based systems in nature. Introduction to necessary biological terminology to describe evolutionary process. Coverage of encoding scheme for mapping problem into genetic systems. Design and implementation of genetic operators such as selection, crossover, mutation and evaluation of fitness.
Particle Swarm Optimisation
Introduction to swarm behaviour in nature and how this can be modelled in computers using populations of embodied agents. Introduction of equations that can be used to drive a swarm towards an optimum goal state. Case study covering design and implementation of particle swarm optimisation algorithm to solve a simple multi variable optimisation problem.
Artificial Life
Introduction to emergent behaviour in nature. Introduction to finite state machines and how complex behaviour can emerge from populations of interacting "simple" agents. Case study on Conway's "Game of Life"
Neural Networks
Introduction to the biological neuron, interaction between neurons in the human brain and how this interaction can account for short term and long term memory. Mapping the biological neuron to a computer based artificial neuron. Introduction to neural network architectures and learning algorithms. Case study covering design and implementation of a neural network to solve a specific problem such as written digit recognition.
| Module Content & Assessment | |
|---|---|
| Assessment Breakdown | % |
| Other Assessment(s) | 40 |
| Formal Examination | 60 |