A hybrid training method for B-spline neural networks

Current and past research has brought up new views related to the optimization of neural networks. For a fixed structure, second order methods are seen as the most promising. From previous works we have shown how second order methods are of easy applicability to a neural network. Namely, we have proved how the Levenberg-Marquard possesses not only better convergence but how it can assure the convergence to a local minima. However, as any gradient-based method, the results obtained depend on the startup point. In this work, a reformulated Evolutionary algorithm - the Bacterial Programming for Levenberg-Marquardt is proposed, as an heuristic which can be used to determine the most suitable starting points, therefore achieving, in most cases, the global optimum.

A microelectrode impedance method to measure interaction of cells

An impedance method was developed to determine how immune system cells (hemocyte) interact with intruder cells (parasites). When the hemocyte cells interact with the parasites, they cause a defensive reaction and the parasites start to aggregate in clusters. The level of aggregation is a measure of the host-parasite interaction, and provides information about the efficiency of the immune system response. The cell aggregation is monitored using a set of microelectrodes. The impedance spectrum is measured between each individual microelectrode and a large reference electrode. As the cells starts to aggregate and settle down towards the microelectrode array the impedance of the system is changed. It is shown that the system impedance is very sensitive to the level of cell aggregation and can be used to monitor in real time the interaction between hemocyte cells and parasites.