Skip to main content
KBS_Icon_questionmark link-ico

 

 

Nature-Inspired Learning Algorithms

 

Key information

  • Module code:

    7CCSMBIM

  • Level:

    7

  • Semester:

      Spring

  • Credit value:

    15

Module description

Aims and Learning Outcomes

Computational methods inspired by evolution, by nature, and by the brain are being used increasingly to solve complex problems in engineering, informatics, robotics and artificial intelligence. They are particularly useful in areas such as optimisation, pattern recognition, scheduling and intelligent control where traditional mathematical and computation methods fail. In such domains, biological ideas have provided valuable models for successful problem solving strategies.

The aim of this module is to introduce students to a range of important biologically-inspired algorithms and techniques and to establish a practical understanding of the real-world problems to which these techniques may fruitfully be applied.

Awareness of biologically-inspired engineering techniques and of the multidisciplinary background to such methods. Appreciation of how new ideas in science and engineering can emerge from lateral thinking and ideas from other disciplines. Understanding of a range of biologically-inspired methods and the ability to apply them to solve real-world problems. Ability to apply genetic algorithms and ant-colony optimisation to solve engineering optimization problems. Ability to recognise future opportunities for exploiting biological inspiration in solving engineering problems.

On successful completion of this module, students will:

Demonstrate:

  • Awareness of biologically-inspired engineering techniques and of the multidisciplinary background to such methods
  • Appreciation of how new ideas in science and engineering can emerge from lateral thinking and ideas from other disciplines
  • Understanding of a range of biologically-inspired methods and the ability to apply them to solve real-world problems
  • Ability to apply genetic algorithms and ant-colony optimisation to solve engineering optimization problems
  • Ability to recognise future opportunities for exploiting biological inspiration in solving engineering problems

Syllabus

  • Concept of optimisation
  • Traditional optimisation methods
  • Evolutionary-Inspired Methods: Binary genetic algorithms, continuous genetic algorithms, permutation genetic algorithm, genetic programming, advanced topics of genetic algorithms, evolutionary strategies
  • Behaviour-Inspired Methods: Ant colony optimisation, particle swarm optimisation
  • Applications of biologically inspired methods

Assessment details

70% examination (3 hours)

30% coursework