Stephen Kelly

Learn More
BACKGROUND Follicular structures resembling germinal centres (GCs) that are characterized by follicular dendritic cell (FDC) networks have long been recognized in chronically inflamed tissues in autoimmune diseases, including the synovium of rheumatoid arthritis (RA). However, it is debated whether these ectopic structures promote autoimmunity and chronic(More)
We investigated whether HSV gene transfer of HSP72 in vivo and in vitro: (1) protected cornu ammonis 1 region of the hippocampus neurons from global cerebral ischemia; and (2) affected Bcl-2 expression. HSV vectors expressing HSP72 and beta-galactosidase (reporter) or beta-galactosidase only (control vector) were injected into cornu ammonis 1 region of the(More)
The serine/threonine kinase, glycogen synthase kinase 3beta (GSK3beta), is abundant in CNS and is neuron specific. GSK3beta plays a pivotal role in the regulation of numerous cellular functions. GSK3beta phosphorylates and thereby regulates many metabolic, signaling, and structural proteins which can influence cell survival. Increased GSK3beta correlates(More)
Half-field Offense (HFO) is a sub-task of Robocup 2D Simulated Soccer. HFO is a challenging, multi-agent machine learning problem in which a team of offense players attempt to manoeuvre the ball past a defending team and around the goalie in order to score. The agent's sensors and actuators are noisy, making the problem highly stochastic and partially(More)
Keepaway soccer is a challenging robot control task that has been widely used as a benchmark for evaluating multi-agent learning systems. The majority of research in this domain has been from the perspective of reinforcement learning (function approximation) and neu-roevolution. One of the challenges under multi-agent tasks such as keepaway is to formulate(More)
Efficient localisation is a highly desirable property for an autonomous navigation system. Weightless neural networks offer a real-time approach to robotics applications by reducing hardware and software requirements for pattern recognition techniques. Such networks offer the potential for objects, structures, routes and locations to be easily identified(More)
This work reports on an approach to direct policy discovery (a form of reinforcement learning) using genetic programming (GP) for the 3 by 3 by 3 Rubik's Cube. Specifically, a synthesis of two approaches is proposed: 1) a previous group theoretic formulation is used to suggest a sequence of objectives for developing solutions to different stages of the(More)