Gaussian fields for semi-supervised regression and correspondence learning

Gaussian fields (GF) have recently received considerable attention for dimension reduction and semi-supervised classification. In this paper we show how the GF framework can be used for semi-supervised regression on high-dimensional data. We propose an active learning strategy based on entropy minimization and a maximum likelihood model selection method. Furthermore, we show how a recent generalization of the LLE algorithm for correspondence learning can be cast into the GF framework, which obviates the need to choose a representation dimensionality.     

Robust seismic design optimization of steel structures

Stochastic performance measures can be taken into account, in structural optimization, using two distinct formulations: robust design optimization (RDO) and reliability-based design optimization (RBDO). According to a RDO formulation, it is desired to obtain solutions insensitive to the uncontrollable parameter variation. In the present study, the solution of a structural robust design problem formulated as a two-objective optimization problem is addressed, where cross-sectional dimensions, material properties and earthquake loading are considered as random variables. Additionally, a two-objective deterministic-based optimization (DBO) problem is also considered. In particular, the DBO and RDO formulations are employed for assessing the Greek national seismic design code for steel structural buildings with respect to the behavioral factor considered. The limit-state-dependent cost is used as a measure of assessment. The stochastic finite element problem is solved using the Monte Carlo Simulation method, while a modified NSGA-II algorithm is employed for solving the two-objective optimization problem.     

Differences in susceptibility to Mannheimia haemolytica-associated mastitis between two breeds of dairy sheep

We used a Mannheimia haemolytica isolate to study differences in susceptibility to experimental mastitis between two breeds of dairy sheep. The isolate was deposited into the teat duct of Karagouniko (K, n=8) or Frisarta (F, n=8) ewes. The animals were monitored by means of clinical, bacteriological, cytological and pathological methods. K ewes did not develop any systemic or mammary clinical signs, whilst F ewes became ill and developed acute clinical mastitis 12 h later (P<0.001). Bacteria were isolated from 34/48 samples from K ewes and from 46/46 samples from F ewes. Positive California mastitis test (CMT) results were 17/24 samples from K ewes and 23/23 samples from F ewes; leucocytes were seen in Giemsa-stained films. Total pathology score summed over all group K ewes was 41 (maximum possible: 128); Man. haemolytica was isolated from 12/24 tissue samples. Total pathology score summed over all group F ewes was 93; Man. haemolytica was isolated from 24/24 tissue samples. Hyperplastic lymphoid nodules consisting of lymphocytes and plasma cells with germinal activity were characteristically present at the border between teat duct-teat cistern of group K ewes; no such structures were observed in teats of group F ewes. The results identified differences in susceptibility/resistance to a mastitis pathogen among animals of the two breeds. Defence mechanisms of the teat appeared to be inadequate against the invading organisms; as lymphoid nodules have been considered important defensive mechanisms of the ovine teat, their observed lack in Frisarta ewes might have predisposed them to development of mastitis.     

Local and global bifurcations in a small power system

This paper presents a number of local and one global (homoclinic loop) bifurcation observed in a small power system involving a load tap changer transformer and an induction motor. Besides the well-known generic local bifurcations (saddle-node and generic Hopf), the paper presents examples of higher codimension bifurcations, such as fold with double zero eigenvalues, degenerate Hopf, and swallowtail bifurcation.     

连接分形的城市

和所有生命系统一样,有活力的城市本质上都具有分形特征。汽车和人口增长带来的压力迫使20世纪的城市规划者推行反分形的几何类型,传统城市的分形特征被去除,给城市肌理造成灾难性的影响。要消除这种破坏,必须深入细致地理解一些问题：（1）分形特征都有哪些;（2）有活力的城市肌理中错综复杂的相互连接;（3）连接和修补城市空间的方法;（4）整合行人、汽车和公共交通的有效方式;（5）如何将实体联系与电子联系结合起来。首先,我们应该消除对分形结构的一些基本误解,然后我会强调层级一致性的本质和重要性。我们可以用分形标度来检验城市形体——作为城市成功的一个前提;另一个独立标度是连通性,应当从拓扑学角度进行研究。用生物系统进化和因特网的知识来讨论尺度分布、逆幂律度尺度法则和“小世界”网络。这些概念表明当代都市一方面向郊区化蔓延发展,另一方面是摩天楼式发展——这种对极端密度的偏好都是病态的。当代都市面临的挑战是如何用最佳方式实现竞争性连接网络的融合。     

Thin infrared imaging systems through multichannel sampling

The size of infrared camera systems can be reduced by collecting low-resolution images in parallel with multiple narrow-aperture lenses rather than collecting a single high-resolution image with one wide-aperture lens. We describe an infrared imaging system that uses a three-by-three lenslet array with an optical system length of 2.3 mm and achieves Rayleigh criteria resolution comparable with a conventional single-lens system with an optical system length of 26 mm. The high-resolution final image generated by this system is reconstructed from the low-resolution images gathered by each lenslet. This is accomplished using superresolution reconstruction algorithms based on linear and nonlinear interpolation algorithms. Two implementations of the ultrathin camera are demonstrated and their performances are compared with that of a conventional infrared camera.     

Electrochemical study of arsenate and water reduction on iron media used for arsenic removal from potable water

Zerovalent iron filings have been proposed as a filter medium for removing As(III) and As(V) compounds from potable water. The removal mechanism involves complex formation of arsenite and arsenate with the iron surface and with iron oxides produced from iron corrosion. There is conflicting evidence in the literature on whether As(V) can be reduced to As(III) by iron filter media. This research uses electrochemical methods to investigate the redox reactions that occur on the surface of zerovalent iron in arsenic solutions. The effect of arsenic on the corrosion rate of zerovalent iron was investigated by analysis of Tafel diagrams for iron wire electrodes in anaerobic solutions with As(V) concentrations between 100 and 20,000 microg/L. As(V) reduction in the absence of surface oxides was investigated by analysis of chronoamperometry profiles for iron wire electrodes in solutions with As(V) concentrations ranging from 10000 to 106 microg/L. The effect of pH on As(V) reduction was investigated by analyses of chronopotentiometry profiles for iron wire electrodes at pH values of 2, 6.5, and 11. For freely corroding iron, the presence of As(III) and As(V) decreased the iron corrosion rate by a factor of 5 as compared to that in a 3 mM CaSO4 blank electrolyte solution. The decrease in corrosion rate was independent of the arsenic concentration and was due to the blocking of cathodic sites for water reduction by arsenic compounds chemisorbed to the iron surface. The chronoamperometry and chronopotentiometry experiments showed that elevated pH and increased As(III) to As(V) ratios near the iron surface decreased the thermodynamic favorability for As(V) reduction. Therefore, reduction of As(V) occurred only at potentials that were significantly below the apparent equilibrium potentials based on bulk solution pH values and As(III) to As(V) ratios. The potentials required to reduce more than 1% of the As(V) to As(III) were below those that are obtainable in freely corroding iron media. This indicates that there will be minimal or no reduction of As(V) in iron media filters under conditions relevant to potable water treatment.     

Side-information-dependent correlation channel estimation in hash-based distributed video coding

In the context of low-cost video encoding, distributed video coding (DVC) has recently emerged as a potential candidate for uplink-oriented applications. This paper builds on a concept of correlation channel (CC) modeling, which expresses the correlation noise as being statistically dependent on the side information (SI). Compared with classical side-information-independent (SII) noise modeling adopted in current DVC solutions, it is theoretically proven that side-information-dependent (SID) modeling improves the Wyner-Ziv coding performance. Anchored in this finding, this paper proposes a novel algorithm for online estimation of the SID CC parameters based on already decoded information. The proposed algorithm enables bit-plane-by-bit-plane successive refinement of the channel estimation leading to progressively improved accuracy. Additionally, the proposed algorithm is included in a novel DVC architecture that employs a competitive hash-based motion estimation technique to generate high-quality SI at the decoder. Experimental results corroborate our theoretical gains and validate the accuracy of the channel estimation algorithm. The performance assessment of the proposed architecture shows remarkable and consistent coding gains over a germane group of state-of-the-art distributed and standard video codecs, even under strenuous conditions, i.e., large groups of pictures and highly irregular motion content.