Maximum Penalized Likelihood Kernel Regression for Fast Adaptation

This paper proposes a nonlinear generalization of the popular maximum-likelihood linear regression (MLLR) adaptation algorithm using kernel methods. The proposed method, called maximum penalized likelihood kernel regression adaptation (MPLKR), applies kernel regression with appropriate regularization to determine the affine model transform in a kernel-induced high-dimensional feature space. Although this is not the first attempt of applying kernel methods to conventional linear adaptation algorithms, unlike most of other kernelized adaptation methods such as kernel eigenvoice or kernel eigen-MLLR, MPLKR has the advantage that it is a convex optimization and its solution is always guaranteed to be globally optimal. In fact, the adapted Gaussian means can be obtained analytically by simply solving a system of linear equations. From the Bayesian perspective, MPLKR can also be considered as the kernel version of maximum a posteriori linear regression (MAPLR) adaptation. Supervised and unsupervised speaker adaptation using MPLKR were evaluated on the Resource Management and Wall Street Journal 5K tasks, respectively, achieving a word error rate reduction of 23.6% and 15.5% respectively over the speaker-independently model.     

Transverse offset loss between two identical noncircular single-mode waveguides

We derive an approximation for the transverse offset loss based on the Laplacian mode-field half-width (MFHW) and far-field second moment for two arbitrary identical noncircular single-mode waveguides. The relation is reduced to the known result for circular-symmetric, single-mode waveguides. Numerical results for symmetrical rectangular waveguides show that the transverse offset loss calculated from the Laplacian MFHW is more accurate than that from the elliptical Gaussian MFHW for small transverse offsets. To our knowledge, this is the first calculation of the errors in the approximate transverse offset loss relations derived from the Laplacian and the elliptical Gaussian MFHW for noncircular optical waveguides.     

The Electrochemical Response of Single Crystalline Copper Nanowires to Atmospheric Air and Aqueous Solution

In this paper, single crystalline copper nanowires (CuNWs) have been electrochemically grown through anodic aluminum oxide template. The environmental stability of the as‐obtained CuNWs in both 40% relative humidity (RH) atmosphere and 0.1 m NaOH aqueous solution has been subsequently studied. In 40% RH atmosphere, a uniform compact Cu₂O layer is formed as a function of exposure time following the logarithmic law and epitaxially covers the CuNW surfaces. It is also found that the oxide layers on CuNWs are sequentially grown when subjected to the cyclic voltammetry measurement in 0.1 m NaOH solution. An epitaxially homogeneous Cu₂O layer is initially formed over the surface of the CuNW substrates by solid‐state reaction (SSR). Subsequently, the conversion of Cu₂O into epitaxial CuO based on the SSR takes place with the increase of applied potential. This CuO layer is partially dissolved in the solution forming Cu(OH)₂, which then redeposited on the CuNW surfaces (i.e., dissolution‐redeposition (DR) process) giving rise to a mixed polycrystalline CuO/Cu(OH)₂ layer. The further increase of applied potential allows the complete oxidation of Cu₂O into CuO to form a dual‐layer structure (i.e., CuO inner layer and Cu(OH)₂ outer layer) with random orientations through an enhanced DR process.     

Effect of Mn substitution on the oxidation/adsorption abilities of iron(III) oxyhydroxides

In this study, divalent manganese ions [Mn(II)] were substituted a part of divalent iron ions [Fe(II)] present in Fe oxyhydroxides to prepare novel composites (Mn@Feox). The composites were prepared by (1) simultaneous hydrolysis of Fe(II) and Mn(II), and (2) rapid oxidation with H₂O₂. The resulting Mn@Feox prepared with different molar ratios of Fe and Mn was characterized and evaluated for their abilities to adsorb arsenic species [As(III) and As(V)] in aqueous solution. X-ray diffraction and field emission transmission electron microscope analyses revealed Mn@Feox has a δ-(Fe_1?x, Mn_x)OOH-like structure with their mineralogical properties resembling those of feroxyhyte (δ-FeOOH). The increase in Mn substitution in Mn@Feox enhanced the oxidative ability to oxidize As(III) to As(V), but it decreased the adsorption capacity for both arsenic species. The optimal Mn/Fe molar ratio that could endow oxidation and magnetic capabilities to the composite without significantly compromising As adsorption capability was determined to be 0.1 (0.1Mn@Feox). The adsorption of As(III) on 0.1Mn@Feox was weakly influenced by pH change while As(V) adsorption showed high dependence on pH, achieving nearly complete removal at pH?<?5.7 but gradual decrease at pH?>?5.7. The adsorption kinetics and isotherms of As(III) and As(V) showed good conformity to pseudo-second-order kinetics model and Freundlich model, respectively.     

The crystal structure of the immunity protein of colicin E7 suggests a possible colicin-interacting surface

The immunity protein of colicin E7 (ImmE7) can bind specifically to the DNase-type colicin E7 and inhibit its bactericidal activity. Here we report the 1.8-angstrom crystal structure of the ImmE7 protein. This is the first x-ray structure determined in the superfamily of colicin immunity proteins. The ImmE7 protein consists of four antiparallel alpha-helices, folded in a topology similar to the architecture of a four-helix bundle structure. A region rich in acidic residues is identified. This negatively charged area has the greatest variability within the family of DNase-type immunity proteins; thus, it seems likely that this area is involved in specific binding to colicin. Based on structural, genetic, and kinetic data, we suggest that all the DNase-type immunity proteins, as well as colicins, share a "homologous-structural framework" and that specific interaction between a colicin and its cognate immunity protein relies upon how well these two proteins' charged residues match on the interaction surface, thus leading to specific immunity of the colicin.     

Automatic 3D model reconstruction of cutting tools from a single camera

Collision detection by machining simulation requires the 3D models of rotating cutters. However, the 3D models of a cutter and holder are not always available. In this paper, a new method is proposed to design an automatic vision-based 3D modeling system, which is able to quickly reconstruct the 3D model of a cutter and holder when they are installed onto the spindle. Only a single camera is mounted on the machine tool to capture the image of the rotating cutter and holder. By viewing the rotating cutter and holder as an object of surface of revolution, the contour of the imaged cutter and holder can be used to reconstruct the 3D model as a stack of circular cross-sections. Then the complete generating function of the cutter and holder can be recovered from the cross-sections. Finally, the 3D model of the cutter is built by rotating the generating function around the spindle axis. The effectiveness and accuracy of the proposed method are verified by experiments on-machine using 12 kinds of cutters and holders, which can satisfy the requirement of collision detection.     

eSeeTrack--visualizing sequential fixation patterns

We introduce eSeeTrack, an eye-tracking visualization prototype that facilitates exploration and comparison of sequential gaze orderings in a static or a dynamic scene. It extends current eye-tracking data visualizations by extracting patterns of sequential gaze orderings, displaying these patterns in a way that does not depend on the number of fixations on a scene, and enabling users to compare patterns from two or more sets of eye-gaze data. Extracting such patterns was very difficult with previous visualization techniques. eSeeTrack combines a timeline and a tree-structured visual representation to embody three aspects of eye-tracking data that users are interested in: duration, frequency and orderings of fixations. We demonstrate the usefulness of eSeeTrack via two case studies on surgical simulation and retail store chain data. We found that eSeeTrack allows ordering of fixations to be rapidly queried, explored and compared. Furthermore, our tool provides an effective and efficient mechanism to determine pattern outliers. This approach can be effective for behavior analysis in a variety of domains that are described at the end of this paper.