Feature reduction is an important issue in pattern recognition. Lower feature dimensionality could reduce the complexity and enhance the generalization ability of classifiers. In this paper we propose a new supervised dimensionality reduction method based on Locally Linear Embedding and Distance Metric Learning. First, in order to increase the interclass separability, a linear discriminant transformation learnt from distance metric learning is used to map the original data points to a new space. Then Locally Linear Embedding is adopted to reduce the dimensionality of data points. This process extends the traditional unsupervised Locally Linear Embedding to supervised scenario in a clear and natural way. In addition, it can also be seen as a general framework for developing new supervised dimensionality reduction algorithms by utilizing corresponding unsupervised methods. Extensive classification experiments performed on some real-world and artificial datasets show that the proposed method can achieve comparable to or even better results over other state-of-the-art dimensionality reduction methods.
Whole cell patch-clamp recordings from GABAergic cells of thalamic reticular nucleus (RTN) in thalamocortical slices made from postnatal day 6 (P6) to 10 (P10) were used to investigate the pattern of rebound bursts (RBs) triggered by an injection of hyperpolarizing current into RTN cells. The number of RBs in the RTN and the overlying Na+/K+ spikes changed in an agedependent manner. The generation of RBs depended largely on the amplitude of the after-hyperpolarizations (AHPs). RB patterns in response to hyperpolarizing current injection into relay cells were markedly different from RB patterns in RTN cells with an after-depolarization. GABAA receptor antagonist bicuculline methiodide (BMI) changed burst firing patterns, increasing the duration of RB and decreasing the amplitude of AHP in RTN cells. Furthermore, local puffs of NMDA in the presence of BMI induced RBs. K+ channel blocker 4-aminopyridine partially mimicked the effect of BMI on AHPs. The shapes of RBs were altered by a selective CaMKII inhibitor KN-62, but not by an inactive analog KN-04., X. Wang ... [et al.]., and Obsahuje bibliografii a bibliografické odkazy