Using the cavity method and diagrammatic methods, we model the dynamics of batch learning of restricted sets of examples, widely applicable to general learning cost functions, and fully taking into account the temporal correlations introduced by the recycling of the examples. The approach is illustrated using the Adaline rule learning teacher-generated or random examples.

No abstract

we developed a simple, low-cost and novel method for constructing three dimensional (3D) microelectrodes in biomedical devices by utilizing low melting point metal alloy. Three dimensional electrodes have unique properties in biomedical applications such as cell lysis, electroporation, dielectrophoresis, and detection. Compared to other fabrication methods for 3D electrodes, the presented one has many advantages, such as normal experiment conditions, simple processes and low-cost equipments. Numerical analysis on electric potential distribution with different electrode configurations was presented to verify the unique field distribution of arc-shape electrodes fabricated by our method. The application of 3D electrode made by high-conductivity metal alloy microspheres was confirmed by continuous particle sorting based on dielectrophoresis (DEP). The proposed technique offers alternatives to construct 3D electrodes from 2D planer electrodes in microfluidic devices.