Evolving a Dynamic Predictive Coding Mechanism for Novelty Detection

Haggett, Simon J. and Chu, Dominique and Marshall, Ian W. (2007) Evolving a Dynamic Predictive Coding Mechanism for Novelty Detection. In: 27th SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence. Springer pp. 167-180. ISBN 978-1-84800-093-3 . (Full text available)

Download (336kB) Preview


Novelty detection is a machine learning technique which identifies new or unknown information in data sets. We present our current work on the construction of a new novelty detector based on a dynamical version of predictive coding. We compare three evolutionary algorithms, a simple genetic algorithm, NEAT and FS-NEAT, for the task of optimising the structure of an illustrative dynamic predictive coding neural network to improve its performance over stimuli from a number of artificially generated visual environments. We find that NEAT performs more reliably than the other two algorithms in this task and evolves the network with the highest fitness. However, both NEAT and FS-NEAT fail to evolve a network with a significantly higher fitness than the best network evolved by the simple genetic algorithm. The best network evolved demonstrates a more consistent performance over a broader range of inputs than the original network. We also examine the robustness of this network to noise and find that it handles low levels reasonably well, but is outperformed by the illustrative network when the level of noise is increased.

Item Type: Conference or workshop item (Paper)
Additional information: To appear in
Uncontrolled keywords: Novelty Detection, Neural Networks, Neuroevolution, Evolutionary Algorithms
Subjects: Q Science > QA Mathematics (inc Computing science) > QA 76 Software, computer programming,
Divisions: Faculties > Sciences > School of Computing > Applied and Interdisciplinary Informatics Group
Depositing User: Dominique Chu
Date Deposited: 24 Nov 2008 18:04 UTC
Last Modified: 08 Jul 2014 08:39 UTC
Resource URI: https://kar.kent.ac.uk/id/eprint/14526 (The current URI for this page, for reference purposes)
  • Depositors only (login required):