Overview
- Students gain an understanding of computational intelligence, an appreciation of its role, and the capabilities of the technologies available.
- Students gain a basic understanding of eight important technologies in computational intelligence:
evolutionary algorithms, neural networks, particle swarm optimisation, ant colony optimisation,
artificial immune systems, learning classifier systems, fuzzy logic, and Bayesian networks.
- Students gain in-depth experience in the use of one of the above technologies.
- Students gain experience in the spoken and written presentation of their ideas to their peers.
The CITS7212 group will be divided into four teams of approximately 2-4 students each.
All students will be expected to gain a basic understanding of all eight CI technologies;
in addition, each team will specialise in two of these technologies, which they will use to implement a solution to a
standard project and as the basis for their presentations to the group.
A significant component of the unit will be finding and reading research papers in the selected areas of CI.
The handbook entry and general unit policies are all available from the
unit outline.
Read and understand this document.
The teams and technologies for 2009 have been set:
- Trent: EAs, LCSs
- Jeffery, Gareth: NNs, Bayesian
- Evgeni, Maria: PSO, Fuzzy
- Robin, Sahan: ACO, AISs
The seminar timetable has also been set, and is included in the schedule below.
Contact Hours
CITS7212 has a lecture at noon-2pm on Mondays in Seminar Room 1.24.
These sessions will vary from week-to-week: some will be traditional lectures,
some will be technical or feedback sessions,
and some will be seminars delivered by students themselves.
CITS7212 also has a lab scheduled at 9-11am on Thursdays. This lab will mostly be unsupervised,
but in some weeks specific events will be scheduled in support of the project.
A six-point unit is a quarter of a full-time workload, so you are expected to commit
10-12 hours/week to CITS7212, averaged over the semester.
Careful attention to time management will ensure that you are not overloaded when deadlines approach.
Assessment
The assessment for CITS7212 consists of
- Three seminars given by each team, worth 10% each.
The first set of seminars will be in Weeks 6-7, on the background of the team's project technology.
The second set of seminars will be in Weeks 8-9, on the background of the team's other technology.
The third set of seminars will be in Weeks 10-11, on applications of the team's project technology, with particular emphasis on the CITS7212 project.
All students will be expected to attend all seminars in the unit.
- A project implemented by each team, worth 40%.
The project will be handed out in Week 5, and it will be due in Week 12.
Each team will construct a solution to a standard problem using their allocated technology.
The team will be required to submit their code and
a report on their project, plus they will give a demonstration of their work to the group in Week 13.
- An exam in November, worth 30%.
The exam will feature one question on each of the eight CI technologies: each student will be expected to answer the
six questions not on their team's technologies.
Schedule
The weekly plan is as follows.
- Nothing scheduled.
- Intro to CITS7212 and to computational intelligence.
- Intro to specific technologies and allocation to teams.
- Consultation with teams.
- Intro to project.
- Teams' seminars on NNs and EAs.
- Teams' seminars on PSO and ACO.
- Teams' seminars on LCSs and Bayesian.
- Teams' seminars on AISs and Fuzzy.
Teams' seminars on applications of NNs and EAs.
-
Monday: teams' seminars on applications of PSO and ACO.
Thursday: teams' seminars on applications of NNs and EAs.
- Project due noon on 16 October.
- Project demos.
|
Administration
Assessment Materials
Other Materials
- Lecture 1: Luigi, Lyndon
- Lecture 2: Luigi, Lyndon
- NNs: paper 1, paper 2
- EAs: paper 1, paper 2
- PSO: tutorial, research paper
- ACO: tutorial, research paper
- LCSs: paper 1, paper 2, paper 3
- AISs: paper 1, paper 2, paper 3
- Intros to Bayesian inference
and Bayesian reasoning
- Intro to fuzzy
Also check out scholarpedia and even wikipedia.
Student Presentations
|
