i++ Departmental Newsletter

Week commencing 10 March 2008

Previous Newsletters

Research News

Essex Robotics research led by Professor Huosheng Hu  has been selected and featured in EXPO21XX's international online e-Exhibitions. 

The EXPO21XX website features several videos of the robotics research being carried out at Essex and film of the robotic fish in action at the London Aquarium. 

BBC4 - The Worlds of Fantasy

Professor Richard Bartle appeared in the final episode of BBC4's series, The Worlds of Fantasy on Wednesday 12th March (and its immediate repeat 3 hours later on Thursday 13th March). He discussed the differences between game worlds and books. "Every time I read The Lord of the Rings, Sam never pushes Frodo into the Crack of Doom. Nobody likes Frodo, they'd love to see Sam push him into the Crack of Doom, but he never does. Why can't Sam, just once, get to push Frodo into the Crack of Doom?"

The episode "Through the Looking Glass" is no longer available on BBC iplayer.

Papers Published/Awaiting Publication

Learning to Recognise Mental Activities: Genetic Programming of Stateful Classifiers for Brain-Computer Interface

Authors: Alexandros Agapitos, Matthew Dyson, Simon M. Lucas and Fransisco Sepulveda

Conference:2008 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE (GECCO-2008) July 12-16, 2008 (Saturday-Wednesday) Atlanta, GA, USA
 
Abstract:
Two families (stateful and stateless) of genetically programmed classifiers were tested on a five class brain-computer interface (BCI) data set of raw EEG signals. The ability of evolved classifiers to discriminate mental tasks from each other were analysed in terms of accuracy, precision and recall. A model describing the dynamics of state usage in stateful programs is introduced.  An investigation of relationships between the model attributes and associated classification results was made.  The results show that both stateful and stateless programs can be successfully evolved for this task, though stateful programs start from lower fitness and take longer to evolve.  

On the Genetic Programming of Time-Series Predictors for Supply Chain Management

Authors: Alexandros Agapitos, Matthew Dyson, Yevgeniya Kovalchuk, Simon M. Lucas

Conference: 2008 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE (GECCO-2008) July 12-16, 2008 (Saturday-Wednesday) Atlanta, GA, USA
 
Abstract:
Single and multi-step time-series predictors were evolved for forecasting minimum bidding prices in a simulated supply chain management scenario. Evolved programs were allowed to use primitives that facilitate the statistical analysis of historical data. 

An investigation of the relationships between the use of such primitives and the induction of both accurate and predictive solutions was made, with the statistics calculated based on three input data transformation methods: integral, differential, and rational.  Results are presented showing which features work best for both single-step and multi-step predictions.

Using genetic algorithm to select the presentation order of training patterns that improves simplified fuzzy ARTMAP classification performance

Authors:R. Palaniappan and C. Eswaran

Applied Soft Computing, DOI: 10.1016/j.asoc.2008.03.003.