The C-loss function for pattern classification

This paper presents a new loss function for neural network classification, inspired by the recently proposed similarity measure called Correntropy. We show that this function essentially behaves like the conventional square loss for samples that are well within the decision boundary and have small errors, and L₀ or counting norm for samples that are outliers or are difficult to classify. Depending on the value of the kernel size parameter, the proposed loss function moves smoothly from convex to non-convex and becomes a close approximation to the misclassification loss (ideal 0–1 loss). We show that the discriminant function obtained by optimizing the proposed loss function in the neighborhood of the ideal 0–1 loss function to train a neural network is immune to overfitting, more robust to outliers, and has consistent and better generalization performance as compared to other commonly used loss functions, even after prolonged training. The results also show that it is a close competitor to the SVM. Since the proposed method is compatible with simple gradient based online learning, it is a practical way of improving the performance of neural network classifiers.     

Towards subject independent continuous sign language recognition: A segment and merge approach

This paper presents a segment-based probabilistic approach to robustly recognize continuous sign language sentences. The recognition strategy is based on a two-layer conditional random field (CRF) model, where the lower layer processes the component channels and provides outputs to the upper layer for sign recognition. The continuously signed sentences are first segmented, and the sub-segments are labeled SIGN or ME (movement epenthesis) by a Bayesian network (BN) which fuses the outputs of independent CRF and support vector machine (SVM) classifiers. The sub-segments labeled as ME are discarded and the remaining SIGN sub-segments are merged and recognized by the two-layer CRF classifier; for this we have proposed a new algorithm based on the semi-Markov CRF decoding scheme. With eight signers, we obtained a recall rate of 95.7% and a precision of 96.6% for unseen samples from seen signers, and a recall rate of 86.6% and a precision of 89.9% for unseen signers.     

Application of an interpenetrating network model to the solid deformation of a quenched isotactic polypropylene film

The molecular orientation and deformation mechanisms of a quenched isotactic polypropylene (iPP) film deformed at temperatures between 303 K and the melting point are studied. At draw temperature T_E less than 400 K where the degree of crystallinity does not change markedly, a linear relationship between molecular orientations of the crystalline and the amorphous phases is revealed and the slope is estimated about 1.82. The interpenetrating network (IPN) model, that takes into account the plastic response of the crystalline (C) network formed by a small portion of crystallites adhered through intercrystalline links and the pseudo-affine deformation of the crystallite enhanced amorphous matrix (CEAM) network, is able to account for inhomogeneous deformation behavior on the mesoscale accompanied with the localized necking in this T_E range. Meanwhile, the initial Young's modulus and the true yield stress exerted by the deformation of the rigid C network exhibit the Arrhenius type of dependence on T_E. The apparent shear modulus of the CEAM network as a function of T_E is discussed in relation to variations in numbers and average molecular weights of the crystalline and the amorphous sequences being manifested by small consecutive endothermic and exothermic peaks in the DSC curve. The IPN model becomes invalid for deformations above T_E=400 K where morphological changes are induced due to melting of crystallites as proved from the DSC measurement.     

Design and evaluation of a high throughput QoS-aware and congestion-aware router architecture for Network-on-Chip

This paper proposes a novel QoS-aware and congestion-aware Network-on-Chip architecture that not only enables quality-oriented network transmission and maintains a feasible implementation cost but also well balance traffic load inside the network to enhance overall throughput. By differentiating application traffic into different service classes, bandwidth allocation is managed accordingly to fulfill QoS requirements. Incorporating with congestion control scheme which consists of dynamic arbitration and adaptive routing path selection, high priority traffic is directed to less congested areas and is given preference to available resources. Simulation results show that average latency of high priority and overall traffic is improved dramatically for various traffic patterns. Cost evaluation results also show that the proposed router architecture requires negligible cost overhead but provides better performance for both advanced mesh NoC platforms.     

Social virtual world continuance among teens: uncovering the moderating role of perceived aggregate network exposure

Engagement in virtual worlds has become pervasive, particularly among the young. At the same time, the number of virtual environments has increased rapidly. Due to intensifying competition, promoting sustained usage, i.e. continuance, has become a top priority for virtual world operators. Prior research has shown that network externalities play a key role in the adoption of communication technologies. However, a small amount of research has examined the role of network externalities in continued IT usage in general or with respect to the virtual world participation in particular. To fill in this gap, we examine how perceived network externalities affect the continuance of social virtual worlds. To this end, we introduce the concept of perceived aggregate network exposure (PANE). We extend the original information systems (IS) continuance model with perceived enjoyment and position PANE as a moderator. We test the model with data collected from 2134 Finnish Habbo Hotel users and employ structural equation modelling in the analysis. The results demonstrate that PANE moderates the influence of motivational factors on continued use intention and satisfaction.     

基于动力有限元分析和神经网络技术的含分层复合材料层合板的损伤诊断

本文基于作者提出的含层间分层损伤层合板的动力有限元分析模型和方法,计算了分层长度和位置对含层间分层损伤层合板结构的固有频率的影响,然后应用MATLAB的神经网络工具箱建立了人工神经网络,通过典型结构的仿真结果比较,证明了采用有限元动力分析和BP网络技术相结合的方法是一种可用于复合材料层合板的分层损伤诊断的有效方法.     

BP网络在加热炉效率评定中的应用

以工艺管式炉有关数据为例，将BP网络有效应用于数据的压缩存储和快速计算，为加热炉效率评定软件提供了BP网络模型。利用MATLAB软件实现了BP网络的训练及仿真分析，并讨论了应用过程中参数的选择以及训练数据的特点对网络训练产生的影响。     

季戊四醇的超分子化学

介绍季戊四醇的超分子化学,包括超分子网状结构、螺环冠醚、非常规液晶、富勒烯衍生物、固相光催化反应、Janus树枝状分子等等。