Integration of prior knowledge of measurement noise in kernel density classification

Samples can be measured with different precisions and reliabilities in different experiments, or even within the same experiment. These varying levels of measurement noise may deteriorate the performance of a pattern recognition system, if not treated with care. Here we seek to investigate the benefit of incorporating prior knowledge about measurement noise into system construction. We propose a kernel density classifier which integrates such prior knowledge. Instead of using an identical kernel for each sample, we transform the prior knowledge into a distinct kernel for each sample. The integration procedure is straightforward and easy to interpret. In addition, we show how to estimate the diverse measurement noise levels in a real world dataset. Compared to the basic methods, the new kernel density classifier can give a significantly better classification performance. As expected, this improvement is more obvious for small sample size datasets and large number of features.     

Minimax Universal Decoding With an Erasure Option

Motivated by applications of rateless coding, decision feedback, and automatic repeat request (ARQ), we study the problem of universal decoding for unknown channels in the presence of an erasure option. Specifically, we harness the competitive minimax methodology developed in earlier studies, in order to derive a universal version of Forney's classical erasure/list decoder, which in the erasure case, optimally trades off between the probability of erasure and the probability of undetected error. The proposed universal erasure decoder guarantees universal achievability of a certain fraction xi of the optimum error exponents of these probabilities (in a sense to be made precise in the sequel). A single-letter expression for xi, which depends solely on the coding rate and the Neyman-Pearson threshold (to be defined), is provided. The example of the binary-symmetric channel is studied in full detail, and some conclusions are drawn     

有线和无线宽带接入产品技术

射频存储系统的目标市场有两个，一个是宽带网络，一个是无线局域网：下图介绍的是第一公里的无线宽带接入系统。图的中央有一个中枢（hub)，它的周围是一个分布网络。这是一个包括光和电的、数据和模拟结合的网络，这种系统在中国的大城市都有广泛的部署。圆圈是飞利浦开发的一些结点、中枢和分布网络，包括放大器和光电设备。     

Boys and violence: A gender-informed analysis.

This article discusses the phenomenon of youth violence from a psychology of gender perspective. Although other factors are discussed--including gun availability, violence-related media influence, family and caretaker factors, and effects of teasing and bullying--the intention is to highlight new thinking on the potential relationship between boys' traditional masculine socialization experiences and violence. In this new perspective, traditional masculine socialization estranges and isolates many boys from their genuine inner lives and vital connections to others, which is theorized to heighten their risk of engaging in acts of violence. The authors identify school and community programs that may be helpful in counteracting damaging socialization experiences and supporting boys' healthier emotional and psychological development. Finally, the article discusses approaches that psychologists and other mental health professionals can use to help address this vital issue. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Universal linear least squares prediction: upper and lower bounds

We consider the problem of sequential linear prediction of real-valued sequences under the square-error loss function. For this problem, a prediction algorithm has been demonstrated whose accumulated squared prediction error, for every bounded sequence, is asymptotically as small as the best fixed linear predictor for that sequence, taken from the class of all linear predictors of a given order p. The redundancy, or excess prediction error above that of the best predictor for that sequence, is upper-bounded by A/sup 2/P ln(n)/n, where n is the data length and the sequence is assumed to be bounded by some A. We provide an alternative proof of this result by connecting it with universal probability assignment. We then show that this predictor is optimal in a min-max sense, by deriving a corresponding lower bound, such that no sequential predictor can ever do better than a redundancy of A/sup 2/p ln(n)/n.     

Gliotoxicity, reverse transcriptase activity and retroviral RNA in monocyte/macrophage culture supernatants from patients with multiple sclerosis

Utilizing cultured lenses from normal and homozygous glutathione peroxidase (GSHPx-1) knockout mice and inhibitors for GSSG Reductase (GSSG Red), 1,3-bis(2-chlorethyl)-1-nitrosourea (BCNU) and catalase (Cat), 3-aminotriazole (3-AT), the ability to degrade H2O2 was examined at two H2O2 concentrations, 300 microM and 80 microM. It was found that GSHPx-1 contributed about 15% to the H2O2 degradation. The Cat contribution was concentration dependent being about 30% at 300 microM H2O2 and approximately 8% to 15% at 80 microM H2O2. GSH loss measured as nonprotein thiol (NP-SH) was shown to be linked to most of the remaining H2O2 degradation accounting for about 54% to 72% of the H2O2 degradation at 300 microM and 80 microM, respectively. However, based on evaluation of the ability of GSH to nonenzymatically degrade H2O2, it can only account for about 36% at 300 microM and 19% at 80 microM H2O2 of the observed lens H2O2 degradation. It is, therefore, concluded that lens GSH must be involved in other reactions either directly or indirectly related to H2O2 degradation.     

Reliable computation by networks in the presence of noise

Lower bounds on the depth of Boolean networks that can compute reliably in the presence of randomly occurring failures are proved. A bound is also given on the reliability that error-tolerant networks can achieve: this bound implies a limit strictly smaller than 1/2 on the failure probability per gate that can be tolerated. The results improve upon recently published bounds of N. Pippenger (ibid., vol.IT-34, p.194-7, 1988) on the depth of error-tolerant formulas and extend the bounds to the case of reliable computation by networks     

Preparation and properties ofgem-dichlorocyclopropane derivatives of long-chain fatty esters

Methyl oleate (18∶1) and linoleate (18∶2) were readily transformed to the correspondinggem-dichlorocyclopropane derivatives in high yield, using triethylbenzylammonium chloride as the phase-transfer catalyst in the presence of aqueous NaOH and CHCl₃. Reaction of dichlorocarbene with methyl 12-hydroxystearate furnished methyl 12-chlorostearate (49%) and 12-O-formylstearate (19%). The hydroxy group in methyl ricinoleate was protected (O-tetrahydropyran-2′-yl) prior to dichlorocyclopropanation of the ethylenic bond. Removal of the protecting group allowed the hydroxy group to be converted to a chloride,O-acetyl, azido orO-formyl function. Treatment of methyl ricinoleate with thionyl chloride, followed by the reaction with dichlorocarbene gave the corresponding 12-chloro-dichlorocyclopropane derivative. The dichlorocyclopropane derivative of oleic acid was transformed to a C₁₉ allenic fatty acid when treated witht-butyl lithium. However, the remaining dichlorocyclopropane derivatives, containing an additional functional group in the alkyl chain, failed to yield the corresponding allenic derivatives. All derivatives were characterized by a combination of spectroscopic and chromatographic techniques, including infrared,¹H nuclear magnetic resonance (NMR), and¹³C NMR spectroscopy.     

A performance- and energy-oriented extended tuning process for time-step-based scientific applications

The Journal of Supercomputing - Scientific application codes are often long-running time- and energy-consuming parallel codes, and the tuning of these methods towards the characteristics of a...