Research on multi-degree-of-freedom neurons with weighted graphs

In this paper, we redefine the sample points set in the feature space from the point of view of weighted graph and propose a new covering model - Multi-Degree-of-Freedorn Neurons (MDFN). Base on this model, we describe a geometric learning algorithm with 3-degree-of-freedom neurons. It identifies the sample points secs topological character in the feature space, which is different from the traditional "separation" method. Experiment results demonstrates the general superiority of this algorithm over the traditional PCA+NN algorithm in terms of efficiency and accuracy.

Research on multi-degree-of-freedom neurons with weighted graphs

In this paper, we redefine the sample points set in the feature space from the point of view of weighted graph and propose a new covering model - Multi-Degree-of-Freedorn Neurons (MDFN). Base on this model, we describe a geometric learning algorithm with 3-degree-of-freedom neurons. It identifies the sample points secs topological character in the feature space, which is different from the traditional "separation" method. Experiment results demonstrates the general superiority of this algorithm over the traditional PCA+NN algorithm in terms of efficiency and accuracy.

InGaAsP-InP Square Microlasers With a Vertex Output Waveguide

InGaAsP-InP square microlasers with a vertex output waveguide are fabricated by planar processes, and the etched sidewalls of the lasers are confined by insulating layer SiO2 and p-electrode TiAu metals. For a square microlaser with a side length of 30 mu m and a 2-mu m-wide output waveguide, a continuous-wave threshold current is 26 mA at room temperature and output power is 0.72 mW at 86 mA. The mode interval of 21 and 7.4 nm is observed for the microlasers with the side length of 10 and 30 mu m, respectively. Finite-difference time-domain (FDTD) simulations indicate that the lasing modes have incident angles of about 45 degrees at the boundaries of the resonator. In addition, square resonators surrounded by air, SiO2-Ti-Au, and SiO2-Au are compared by FDTD simulations.