Convergence analysis of convex incremental neural networks

Recently, a convex incremental algorithm (CI-ELM) has been proposed in Huang and Chen (Neurocomputing 70:3056–3062, 2007), which randomly chooses hidden neurons and then analytically determines the output weights connecting with the hidden layer and the output layer. Though hidden neurons are generated randomly, the network constructed by CI-ELM is still based on the principle of universal approximation. The random approximation theory breaks through the limitation of most conventional theories, eliminating the need for tuning hidden neurons. However, due to the random characteristic, some of the neurons contribute little to decrease the residual error, which eventually increase the complexity and computation of neural networks. Thus, CI-ELM cannot precisely give out its convergence rate. Based on Lee’s results (Lee et al., IEEE Trans Inf Theory 42(6):2118–2132, 1996), we first show the convergence rate of a maximum CI-ELM, and then systematically analyze the convergence rate of an enhanced CI-ELM. Different from CI-ELM, the hidden neurons of the two algorithms are chosen by following the maximum or optimality principle under the same structure as CI-ELM. Further, the proof process also demonstrates that our algorithms achieve smaller residual errors than CI-ELM. Since the proposed neural networks remove these “useless” neurons, they improve the efficiency of neural networks. The experimental results on benchmark regression problems will support our conclusions. The work is under the funding of Singapore MOE AcRF Tier 1 grant WBS No: R 252-000-221-112.     

改进的RBF神经网络在磨矿指标预测中的应用

针对选矿生产过程中磨矿生产率和磨矿浓度、介质填充率和料球比3个指标之间的关系,分析了磨矿生产率对磨矿过程产品质量和工作效率的影响,提出了磨矿过程预测模型,采用了模糊聚类与改进的遗传算法相结合训练RBF神经网络的新方法。仿真结果表明,该模型可以快速准确地预报生产状况,对于企业进行实时生产控制具有较高的实用价值。     

Online face recognition system based on multi-view algorithm and Kernel PCA

An online face recognition system is presented in the paper. To online face recognition system, we should consider the recognition rate, the image compression and image size. In the paper we researched the innovation technologies for face recognition system, including Kernel Principal Component Analysis （Kernel PCA）, Delta low-pass wavelet filter, and face recognition algorithm using multiple images. Kernel PCA is derived to classify the characteristics of training images in the database. Delta low-pass wavelet filter is used to reduce the image size. A face recognition algorithm using multiple images is presented to improve the recognition rate. Simulation experiment shows that in the case of packet loss recognition rate is improved highly.     

MCLP-based methods for improving “Bad” catching rate in credit cardholder behavior analysis

Cardholders’ behavior prediction is an important issue in credit card portfolio management. As a promising data mining approach, multiple criteria programming (MCLP) has been successfully applied to classify credit cardholders’ behavior into two groups. In order to better control credit risk for financial institutes, this paper proposes three methods based on MCLP to improve the “Bad” catching accuracy rate. One is called MCLP with unbalanced training set selection, the second is called fuzzy linear programming (FLP) method with moving boundary, and the third is called penalized multi criteria linear programming (PMCLP). The experimental examples demonstrate the promising performance of these methods.     

Effects of shear rate and small periodic corrugation on the slip velocity in microscopic domain

We obtain analytically the slip velocity (up to the second order) of shear-thinning fluids inside the periodically corrugated microtube by using the verified fluid model and boundary perturbation method. Our results show that, even the slip length being zero, there is a slip velocity which is proportional to the small amplitude of periodic corrugation, the (referenced) shear rate, the applied forcing, and the orientation or the angle. Our results could be applied to the flow control in microfluidics as well as biofluidics.     

Classification of metabolic and respiratory demands in fire fighting activity with extreme workloads

Fire fighting work comprises work tasks requiring an energy yield at maximal or close to maximal levels of the individual. Due to the very nature of fire fighting more complex physiological variables are difficult to measure. We measured metabolic and respiratory responses in 15 male, professional fire fighters during simulated work tasks on a test ground. Work time was on the average 22 min with individual components of work tasks lasting 2-4 min. The mean oxygen consumption for the whole exercise (22 min) was 2.75+/-0.29 l/min. The most demanding work task demanded an oxygen uptake of 3.55+/-0.27 l/min. Corresponding values for respiratory minute volumes were 82+/-14 and 102+/-14l/min, respectively. Heart rates averaged 168+/-12 for the whole test and 179+/-13 beats/min for the heaviest work task. Two new classes for classification of intensive and exhausting, short term physical work are proposed for inclusion in ISO8996 and values for relevant parameters are proposed.     

静态图像编码研究进展

多媒体及Internet的迅速发展导致了对图像编码技术更高的要求。首先以波形编码、第二代编码技术和分形编码技术为线索对静态图像编码技术的发展及编码技术中所存在的一些问题进行了讨论,并从图像编解码的观点讨论了图像在有噪信道上传输的误码率情况和适应于传输的一些图像编码技术;同时对即将成为新的静态图像编码标准的JPEG－2000的目标、编码机制、POI编码的Maxshift方法以及对码流的直接操作等情况进行了讨论。尽管压缩效果一直是任何编码方案所追求的目标,但其它特性,像基于内容的编码、可分级性编码以及鲁棒性编码等也是选择一个好的编码方案时应该考虑的,据此最后对图像编码技术的发展方向进行了展望。     

Hopfield型连续神经网络指数收敛速度的估计

讨论了Hopfield型连续神经网络的指数收敛速度问题,从理论上得到了新的局部指数收敛条件,与以往的结果相比,本文导出指数收敛的阶更大,指数收敛的速度要更快,文中还给出了数值举例。     

（RbCl）108离子簇均相成核的分子动力学研究

分子动力学模拟测得熔融（RbCl)108离子簇在300K至600K温度时凝固的成核速率大于10^35m^-3s^-1,用均相成核的经典理论加以分析,估算得氯化铷离子簇的固液界面自由能σalT^1.4,用Grunasy提出的扩散界面理论加以分析,估算得扩展界面厚度约为0．195nm,σalT^0.97,2个理论在实验温度范围内没有明显判别,且均能较好地分子动力学模型的结果,但预测的较高温度下的成核速率有比较显著区别,因此进一步地研究将有助于鉴别它们。