Role
Stage tutor for stage 2:
BSc (Hons) Computing.
BSc (Hons) Computer Science
BSc (Hons) Computing and Games Development
Member of Faculty of Science and Technology Ethics Committee.
Qualifications & background
BA (Hons) Psychology
BSc (Hons, 1st) Computing with Design
MSc (Distinction) Intelligent Systems for Engineering Design
Professional membership
Teaching interests
Module Leader:
SOFT141: Network Programming 1
SOFT144: Software Engineering for Games Development
AINT206: Artificial Intelligence for Games Development
Co-teach:
AINT301: Evolutionary Computation and Artificial Life
Previous Modules:
SOFT234: Network Programming II
SOFT221: Server-side Programming and Development
SOFT232: Software Development for Digital Arts
AINT203: An Introduction to Artificial Intelligence
SOFT504: Programming for Intelligent Interactive Systems
ISAD503: Human Computer Interaction
Research interests
Previously my research was centred on the use of Evolutionary Computation and associated Computational Intelligence techniques in Engineering and Architectural design. The main focus of the research is how such techniques can be applied to the conceptual stage of the design process to enhance innovation and creativity on the part of the designer.
More recently I have registered for a PhD in the area of Situated and Embodied Artificial Intelligence. Currently the working title of the PhD is:
'Modelling Modular Neural Development and Compositional Task Handling in Embodied Agents'
The overall aim of the research is to investigate hierarchical modular neural network architectures and their ability to adapt to, and generate behaviours for, compositional / modular type tasks in cognitive robotics. Within this, consideration will also be given to the role of Evolutionary Computation (EC) in evolving such architectures.
Starting with an existing recurrent neural architecture, initial work will be to develop and enhance this network using evolutionary based approaches using a simulated robot. In this instance only simple behavioural primitives (i.e. grasp, lift) will be used. This model will be extended to a hierarchical model allowing the behavioural primitives to be sequenced into a goal directed task. The tasks will involve both sequential (i.e. box stacking) and compositional actions (i.e. nested cups).
The PhD emelemt of the research will examine the ‘generalisation’ capabilities of the existing learned behavioural sequences. Two categories are of interest: Firstly those tasks which require a ‘generalisation’ of existing leaned behaviours to a similar goal related task; secondly those where behaviours are not related to a particular goal, (i.e. tool use). The PhD research will also involve use of EC for the evolution of adaptive, modural neural architecture.
UoP Research group membership
Centre for Robotics and Neural Systems (CRNS)
Other research
Involved with the Arch-OS project
http://www.arch-os.com/
Publications
Book Chapters
Rafiq, Y., Beck, M., Packham, I and Denham S., (2005) Evolutionary
Computation and Visualisation as Decision Support tools for Conceptual
Building Design. (Invited paper) In Topping, B. H. V (Ed) Innovation In Civil and Structural Engineering Computing, chapter 3, pp 49-74, Saxe-Coburg Publications, http://www.saxe-coburg.co.uk/pubs/descrip/sl2005.htm
Parmee, I. C. and Beck, M. A. (1997) An Evolutionary, Agent-Assisted Strategy For Conceptual Design Space Decomposition. In D. Corne and J. L. Shapiro (eds.), Evolutionary Computing: Selected Papers from the 1997 AISB International Workshop, Springer Lecture Notes in Computer Science. .
Journals
De Wilde, P., Rafiq, Y., and Beck, M. (2008) Uncertainties in Predicting
the Impact of Climate Change on Thermal Performance of Domestic Buildings in
the UK. Building Services Engineering Research & Technology, 29 (1),
7-26
Rafiq, Y, and Beck, M. (in press) A Decision Support Tool for Multi-Disciplinary Conceptual Design. The Structural Engineer
Conferences
Phillips, M., Beck, M., Bugman, G., and Speed, C. (2007) The Arch-OS System. IBPSA England Symposium Plymouth, UK.
De Wilde, P., Beck, M., and Rafiq, Y. (2007) Development of a Framework for the Assessment of Climate Change on the Thermal Performance of Buildings. IBPSA England Symposium Plymouth, UK.
De Wilde, P., Beck, M., and Rafiq, Y. (2007) Simulating the Impact of Long Term Change Scenarios on Thermal Performance of Buildings. 10th IBPSA Conference Beijing, China
Rafiq, Y., Beck, M., and De Wilde, P. (2007) How Visualisation Enhances Better Understanding of Thermal Performance of Domestic Buildings. 14th European Group for Intelligent Computing in Engineering (EG-ICE) Workshop
Rafiq, Y., Beck, M., and Packham, I. (2006) Interactive Visualisation Systems for Conceptual Building Design: A Practical Approach. International Conference on Advances in Engineering Structures, Mechanics and Construction, University of Waterloo, Canada.
Rafiq, Y, Packham, I and Beck, M. (2005) Importance of Visualisation in Decision Making Process in Engineering Design. International Conference on Computing in Civil Engineering (ASCE2005) Cancun, Mexico.
Rafiq, Y., Ekwe, A., Beck, M., and Wigginton, M. (2005) Optimising Intelligent Facades using a Genetic Algorithm. In Keiller, A., Ledbetter, S and Wilkinson, M (Eds.) Glass in Buildings 2, University of Bath.
Beck, M. A. and Parmee, I. C. (1999) Design Exploration: Extending the Bounds of the Search Space. In Proceedings of IEEE Congress on Evolutionary Computation (CEC),Washington D.C.
Beck, M. A. and Parmee, I. C. (1999) Extending the Bounds of the Search Space: A Multi-Population Approach. In Proceedings of the Genetic and Evolutionary Computation Conference, GECCO-99, Orlando Florida.
Beck, M. A. and Parmee, I. C. (1999) A Hybrid Adaptive Search Technique for Design Space exploration. In Proceedings of the International Conference on Engineering Design (ICED99) Munich, Germany.
Beck, M. A. and Parmee, I. C. (1997) An Evolutionary Approach to Conceptual Design Exploration. Workshop on Strategic Knowledge and Concept Formation, Loughborough University.
