Introduction

This page details the third year individual projects that are/have been supervised by Graham Kendall who is a Professor in the School of Computer Science at The University of Nottingham.

This page contains details of projects supervised in the academic year 2006/07.

The main project page can be found here.

Downloadable Dissertations

Some of the projects are available for download, so that you can view them. This facility is for current third year students, being supervised by me, who can use previous dissertations for their own research and also to look at previous dissertations to see how to lay them out etc. (it also saves the inevitable requests I receive asking to see a "good" project from last year, with the risk that I never see it again!).

As such, these files are password protected and, to get the username/password you need to EMAIL me. People outside of the university will not be given access.

In addition, I cannot give access to those students outside of my tutor group. You should approach your own tutor to get copies of previous dissertations they have supervised.

Note : I cannot make the files available in formats other than those already supplied.

3rd Year Projects Supervised in 2006/07

Application of neural networks for opponent modelling in poker by Iain Selvage (pdf : 212KB)
Abstract : Opponent modelling is the practice of observing opponents behaviour to predict their decisions in any given situation. In the game of poker the ability to predict an opponent’s next decision gives a player a significant advantage over the competition. Currently high class poker players often adopt a mix of strategies changing strategy based on their position in the game or to mislead their opponents. The ability to track players changing strategy and predict their actions is needed to avoid opponents exploiting weakness in the program and to find weakness in the opponent. The problem of accurately modelling a dynamic player is one that has eluded current research and is seen to be the next step in improving the current ability of poker programs. The problem requires the ability to model adaptively in a noisy domain with incomplete information. To solve the task of opponent modelling within the game of poker the problem is abstracted to one of pattern recognition. Game context will be transformed using a multi layer perceptron into one of three betting decisions open to a player. Methods will be used to optimize the perceptrons ability to generalise including implementing a committee of networks to investigate the level of adaptive ness which can be incorporated into the modeller.