Text Box: GP-ZIP is a system that aims to learn from the given stream of data and automatically evolve Universal Data Compression algorithm.
Text Box: This research proposes an application based on the use of Genetic Programming to identify the files contents by analysing only the raw binary streams and without the need for any meta data.
Text Box: In this research we investigate the idea of predicting localised muscle fatigue by identifying a transition state which resides between the non-fatigue and the fatigue stages within the EMG signal.

Genetic Programming—zip (Completed 2010)

GP– Fileprint (Completed 2010)

GP-Muscle Fatigue  Predictor (Completed 2009)


Text Box: We use Genetic Programming (GP) to generate programs that predict data compression ratio for compression algorithms. GP evolves programs with multiple components. Experiments with the proposed approach show that GP is able to accurately estimate the compression ratio of unseen files without the need to run every compression algorithm in question.

GP– Predictor of Data Compression saving (Completed 2009)

This research proposes a new framework based on the Genetic Programming (GP) to automatically decompose problems into smaller and simpler tasks.  The proposed framework has been tested with several symbolic regression problems. Experimentation reveals that performance of systems employing this framework is significantly outperforming standard GP.

GP for Problem Decomposing (Completed 2010)

This project is associated with the EPSRC project ‘Designing Mechanisms for Automated Resource Allocation’. Research on this project lies at the intersection of computer science and game theory. Specifically, it aims to investigate computational aspects of economic mechanisms such as bargaining, auctions, and coalitions, and develop mechanisms for resource allocation in multi-agent systems.

·  More details...

Automatic Resource Allocation (Completed 2011)

When the optimisation problem gives rise to objective functions which are prohibitively expensive to evaluate (like in aerodynamic simulation), a single optimisation case can take many minutes, hours, or even days to complete and often the whole optimisation process become infeasible. Optimisation methods based on surrogate models, have been successfully employed to tackle expensive objective functions. An attractive way to reduce the search time of EA’s when dealing with expensive objective function is to use a cheap approximation model, that can rank the population similarly as the original expensive evaluation function.


·  More details...

Surrogate Optimisation (ongoing)

Calul d'interactions pales tourbillons

This research proposes an interactive GA system to enhance brain MRI images referred to as SIGA. SIGA aims at aiding neurologists and radiologists to detect abnormalities in brain images. Unlike standard IGA systems where its learning is completely dependant on the user, SIGA enhances the MRI based on a joint decision between user and computer. Also, it uses RBFN as a surrogate to model the user’s preferences and reduces the total number of user evaluations.

·  More details...

SIGA  (ongoing)

RE attempts to continually recover evolvability loss caused by the canonical EA iteration process. It borrows the term “Recurrent” from the taxonomy of Neural Networks (NN), in which a Recurrent NN (RNN) is a special type of network that uses a feedback loop, usually to account for temporal information embedded in the sequence of data points presented to the network. Unlike RNN, the “temporal” dimension in our algorithm pertains to the sequential nature of the evolution process itself; and not to the data sampled from the problem solution space.  This is done by a feedback loop that recurrently adjusts the fitness values of individuals in population ti based on the fitness of their offspring in population ti+1

·  More details...

Recurrent Evolution (RE)  (ongoing)

Computing and Electronic Engineering

Ahmed Kattan

Research project for the Saudi Ministry of Labor to work on improving the jobs nationalisation process and study the Saudi labor market behavior.


·  More details...

Nitaqat Optimisation (ongoing)