i++ School Newsletter

Week commencing 24 January 2011

Previous Newsletters

CSEE Postgraduate Recruitment Event on 1 February 2011

Do you want to continue with your studies/education after graduating?  The CSEE department at the University of Essex offers an excellent and diverse range of Master and PhD schemes.  If your currently a student with us, why not come and find out more at our Postgraduate Information Event taking place on Tuesday 1 February at 9.00 in room 1N1.4.1.  Please contact the CSEE administration team if you would like any further information.

Visiting Fellows join CCFEA

The University has accorded the title of Visiting Fellow to the following scholars from 15 January 2011 – 14 January 2014 whilst they are associated with CCFEA, at the School of Computer Science and Electronic Engineering.

Ms Alma Lilia García Almanza

Award of Title as a Visiting Fellow granted on 15 January 2011 in recognition of her continuing contribution to the School of Computer Science and Electronic Engineering and collaboration with Professor Edward Tsang, Director of CCFEA.
Alma is now working at the Mexican Central Bank, Mexico.

Dr Biliana Alexandrova Kabadjova

Award of Title as a Visiting Fellow granted on 15 January 2011 in recognition of her continuing contribution to the School of Computer Science and Electronic Engineering and collaboration with Professor Edward Tsang, Director of CCFEA.
Biliana is now working at the Mexican Central Bank, Mexico.

Dr Serafin Martinez Jaramillo

Award of Title as a Visiting Fellow granted on 15 January 2011 in recognition of his continuing contribution to the School of Computer Science and Electronic Engineering and collaboration with Professor Edward Tsang, Director of CCFEA.
Serafin is now working at the Mexican Central Bank, Mexico.

Paper published

Al-Mulla, Mohamed R., Francisco Sepulveda and Martin Colley, Evolved pseudo-wavelet function to optimally decompose sEMG for automated classification of localized muscle fatigue, Medical Engineering & Physics, 2010

Abstract - The purpose of this study was to develop an algorithm for automated muscle fatigue detection in sports related scenarios. Surface electromyography (sEMG) of the biceps muscle was recorded from ten subjects performing semi-isometric (i.e., attempted isometric) contraction until fatigue. For training and testing purposes, the signals were labelled in two classes (Non-Fatigue and Fatigue), with the labelling being determined by a fuzzy classifier using elbow angle and its standard deviation as inputs. A genetic algorithm was used for evolving a pseudo-wavelet function for optimising the detection of muscle fatigue on any unseen sEMG signals. Tuning of the generalised evolved pseudo-wavelet function was based on the decomposition of twenty sEMG trials. After completing twenty independent pseudo-wavelet evolution runs, the best run was selected and then tested on ten previously unseen sEMG trials to measure the classification performance. Results show that an evolved pseudo-wavelet improved the classification of muscle fatigue between 7.31% and 13.15% when compared to other wavelet functions, giving an average correct classification of 88.41%.

Keywords: Genetic algorithms; Localized muscle fatigue; sEMG; Wavelet analysis

Papers accepted

Al-Mulla, Mohamed R., Francisco Sepulveda and Martin Colley, An Autonomous Wearable System for Predicting and Detecting Localised Muscle Fatigue, MDPI Sensors

Abstract - Muscle fatigue is an established area of research and various types of muscle fatigue have been clinically investigated in order to fully understand the condition. This paper demonstrates a non-invasive technique used to automate the fatigue detection and prediction process. The system utilises the clinical aspects such as kinematics and surface electromyography (sEMG) of an athlete during isometric contractions. Various signal analysis methods are used illustrating their applicability in real-time settings. This demonstrated system can be used in sports scenarios to promote muscle growth/performance or prevent injury. To date, research on localised muscle fatigue focuses on the clinical side and lacks the implementation for detecting/predicting localised muscle fatigue using an autonomous system. Results show that automating the process of localised muscle fatigue detection/prediction is promising. The autonomous fatigue system was tested on five individuals showing 90.37% accuracy on average of correct classification and an error of 4.35% in predicting the time to when fatigue will onset.

Kampouridis, M., Chen, S.-H., Tsang, E., Market Microstructure: Can Dinosaurs Return? A Self-Organizing Map Approach Under an Evolutionary Framework, Evo*, 27-29 April, Turin, Italy.

Abstract - This paper extends a previous market microstructure model, which investigated fraction dynamics of trading strategies. Our model consisted of two parts: Genetic Programming, which acted as an inference engine for trading rules, and Self-Organizing Maps (SOM), which was used for clustering the above rules into trading strategy types. However, for the purposes of the experiments of our previous work, we needed to make the assumption that SOM maps, and thus strategy types, remained the same over time. Nevertheless, this assumption could be considered as strict, and even unrealistic. In this paper, we relax this assumption. This offers a significant extension to our model, because it makes it more realistic. In addition, this extension allows us to investigate the dynamics of market behaviour. We are interested in examining whether financial markets' behaviour is non-stationary, because this implies that strategies from the past cannot be applied to future time periods, unless they have co-evolved with the market. The results on an empirical financial market show that its behaviour constantly changes; thus, agents' strategies need to continuously adapt to the changes taking place in the market, in order to remain effective.

Kampouridis, M., Chen, S.-H., Tsang, E., Investigating the Effect of Different GP Algorithms on the Non-Stationary Behavior of Financial Markets, IEEE Symposium Series on Computational Intelligence, 11-15 April, Paris, France.

Abstract - This paper extends a previous market microstructure model, where we used Genetic Programming (GP) as an inference engine for trading rules, and Self Organizing Maps as a clustering machine for those rules. Experiments in that work took place under a single financial market and investigated whether its behaviour is non-stationary or cyclic. Results showed that the market's behaviour was constantly changing and strategies that would not adapt to these changes, would become obsolete, and their performance would thus decrease over time. However, because experiments in that work were based on a specific GP algorithm, we are interested in this paper to prove that those results are independent of the choice of such algorithms. We thus repeat our previous tests under two more GP frameworks. In addition, while our previous work surveyed only a single market, in this paper we run tests under 10 markets, for generalization purposes. Finally, we deepen our analysis and investigate whether the performance of strategies, which have not co-evolved with the market, follows a continuous decrease, as it has been previously suggested in the agent-based artificial stock market literature. Results show that our previous results are not sensitive to the choice of GP. Strategies that do not co-evolve with the market, become ineffective. However, we do not find evidence for a continuous performance decrease of these strategies.

Forthcoming Seminars

Wednesday 26 January, 16.00, 1N1.4.1

Problems and Challenges in Evolutionary Dynamic Optimisation

Philipp Rohlfshagen, University of Essex

Abstract - In this talk, I will outline some of the problems and challenges faced in the field of evolutionary dynamic optimisation, a branch of evolutionary computation concerned with the study and application of evolutionary algorithms to the class of dynamic optimisation problems.

Following a brief review of the problem domain and existing techniques, I will highlight some of the pathologies that existing approaches have suffered from. I propose that these problems are largely due to a lack of appreciation for the problem's dynamics and will show how Solution Concepts may be used to address these issues. Finally, I would like to denote some time to give an overview of the work I am currently involved in: Monte Carlo Tree Search for real-time video games.

Optoelectronic Seminar Series

The Optoelectronic Seminar Series continues this term. These seminars will be  presented by professors, researchers and second and third-year PhD students in CSEE, focusing on their activities in nonlinear and quantum optics, novel semiconductor materials, and optoelectronic devices.

They will take place every Wednesday from 1.00 - 2.00pm in room 4.311.

26 January, Ben Royall, GaInNAs MQW and n-i-p-i solar cells.