Research in neurophysiology includes both analysis of data from neuronal systems (networks of brain cells; both live and cultured), and development of models to explain both the processes that form the character of data, and the high level function that these express; i.e. behaviour and thought. Capturing and analysing data from neuronal systems is time-consuming, difficult and expensive: many techniques exist, some using multichannel electrical recording, and some using ion-sensitive fluorescent dyes. Different techniques have different advantages: some have high time resolution, whereas others have high space resolution. The models that derive from this data also exist at many levels, from the detailed modelling of membrane-embedded ion channels and neurotransmitters to compartmental neural models, through models of small neural networks, to larger models of many thousands of neurons. All models and algorithms are hungry for data to determine their many parameters and characteristics. Currently this activity is largely a one-lab science: datasets are shared within a lab, and with some computational modellers. The research is also not organised to ensure that data and models produced by small communities of specialist researchs can easily be integrated to contribute to the bigger picture. Datasets are discarded after the experimentor has completed their experimental report, or are archived in a format that is not widely accessible. This project aims to use the GRID to change that: it will enable experimenters to archive their datasets in a structure, making them widely accessible for modellers and algorithm developers to exploit. Experimental datasets are useless without accurate descriptions of the experimental conditions, and hence an appropriate set of metadata will developed to augment the data, allowing the project researchers to collaborate more widely and persistently by sharing data in a sensible, referenced form. Further, the project will provide integrated and co-ordinated services for the neuroscience data, enabling neuronal signal detection, sorting and analysis, as well as visualisation and modelling. Data security is critically important to experimentors: they do not wish to be simply anonymous contributors of data, but to be directly involved in further analysis of their datasets, and this will be supported. Further we will enable direct near real-time analysis of streamed experimental data, providing information to distributed teams of specialists that will allow difficult experiments to be optimised. These interventions will catalyse a step change in research practice in this area of neuroscience, which will allow best value to be derived from the significant research investment that is made in order to understand the brain.

The field of Machine Learning plays an increasingly important role in Computer Science and related disciplines. Over the past decade, the availability of powerful desktop computers has opened the door to synergistic interactions between empirical and theoretical studies of Machine Learning, showing thevalue of the ``learning from example'' paradigm in a wide variety of applications. Much effort has been devoted by Machine Learning researchers to the standard single task learning problem and exciting results have been derived. However, Machine Learning capabilities are still extremely limited when compared to those of humans. The human ability to generalise knowledge learned in one task in order to solve a new task is not available in current Machine Learning systems. Multi-task learning research has not yet received sufficient attention in the field. The standard single task learning approach builds on assumptions that are too restrictive to be easily extended to the novel learning scenarios which are envisaged in this proposal. Although interesting insights on multi-task learning have been provided, at present there is no comprehensive framework for multi-task learning and no cornerstone has yet been placed in the field. Thus, the main purpose of this proposal is to develop this area of Machine Learning research. The proposal focuses on Statistical Machine Learning methods for learning multiple related (classification or regression) tasks and integrating information across them. We shall design formal models of relationships between the tasks and develop (learning algorithms) for learning these relationships from data. We shall also develop the mathematical foundations (generalisation bounds, approximation results, convergence results) for multi-task learning, extending some key theoretical results for single tasklearning. Furthermore, the learning algorithms will be applied to two key applications, namely user preference modelling and multiple microarray gene expression data analysis. A central role in our approach is played by certain graph structures which allow us to model task relationships. This approach is very general and can be adapted to increasingly complex learning scenarios. The computational methods are based on the minimisation of certain penalty functionals via a large number of hyper-parameters associated with the tasks. The proposed research will lead to a new generation of trainable machines for multi-task learning, which will be more powerful and flexible models of learning, closer to human learning than previously developed Machine Learning frameworks.