Efficient Leave-<Emphasis Type="Italic">m</Emphasis>-out Cross-Validation of Support Vector Regression by Generalizing Decremental Algorithm

We propose a computationally efficient method for cross-validation of the Support Vector Regression (SVR) by generalizing the decremental algorithm of SVR. Incremental and decremental algorithm of Support Vector Machines (SVM) ^{2, 8, 9)} efficiently update the trained SVM model when a single data point is added to or removed from the training set. The computational cost of leave-one-out cross-validation can be reduced using the decremental algorithm. However, when we perform leave-m-out cross-validation (m > 1), we have to repeatedly apply the decremental algorithm for each data point. In this paper, we extend the decremental algorithm of SVR^{8, 9)} in such a way that several data points can be removed more efficiently. Experimental results indicate that the proposed approach can reduce the computational cost. In particular, we observed that the number of breakpoints, which is the main computational cost of the involved path-following, were reduced from \({\mathcal O}(m)\) to \({\mathcal O}(\sqrt{m})\).     

Super-vector coding features extracted from both depth buffer and view-normal-angle images for part-based 3D shape retrieval

We have witnessed 3D shape models abundant in many application fields including 3D CAD/CAM, augmented/mixed reality (AR/MR), and entertainment. Creating 3D shape models from scratch is still very expensive. Efficient and accurate methods for shape retrieval is essential for 3D shape models to be reused. To retrieve similar 3D shape models, one must provide an arbitrary 3D shape as a query. Most of the research on 3D shape retrieval has been conducted with a “whole” shape as a query (aka whole-to-whole shape retrieval), while a “part” shape (aka part-to-whole shape retrieval) is more practically requested as a query especially by mechanical engineering with 3D CAD/CAM applications. A “part” shape is naturally constructed by a 3D range scanner as an input device. In this paper, we focus on the efficient method for part-to-whole shape retrieval where the “part” shape is assumed to be given by a 3D range scanner. Specifically, we propose a Super-Vector coding feature with SURF local features extracted from the View-Normal-Angle image, or the image synthesized by taking account of the angle between the view vector and the surface normal vector, together with the depth-buffered image, for part-to-whole shape retrieval. In addition, we propose a weighted whole-to-whole re-ranking method taking advantage of global information based on the result of part-to-whole shape retrieval. Through experiments we demonstrate that our proposed method outperforms the previous methods with or without re-ranking.     

Compact electrostatic levitator for diffraction measurements with a two axis diffractometer and a laboratory x-ray source

A compact electrostatic levitator was developed for the structural analysis of high-temperature liquids by x-ray diffraction methods. The size of the levitator was 200 mm in diameter and 200 mm in height and can be set up on a two axis diffractometer with a laboratory x-ray source, which is very convenient in performing structural measurements of high-temperature liquids. In particular, since the laboratory x-ray source allows a great amount of user time, preliminary or challenging experiments can be performed with trial and error, which prepares and complements synchrotron x-ray experiments. The present small apparatus also provides the advantage of portability and facility of setting. To demonstrate the capability of this electrostatic levitator, the static structure factors of alumina and silicon samples in their liquid phases were successfully measured.     

Accumulation of triosephosphate isomerase, with sequence homology to Beta amyloid peptides, in vessel walls of the newborn piglet hippocampus

We investigated whether beta-amyloid (Abeta)-like immunoreactivity was seen in the brains of newborn piglets. The immunoreactivity for Abeta(1-42) and Abeta(1-40) proteins, but not Abeta precursor protein, was present in CD68-positive perivascular cells of the hippocampus and in parts of the meninges. It was colocalized with immunoreactivity for receptor for advanced glycation end product and tumor necrosis factor-alpha. The protein with a molecular mass of 27 kDa, which was recognized by the Abeta antibodies, was identified as triosephosphate isomerase (TPI) with sequence homology to Abeta peptides by N-terminal amino acid sequencing, mass fingerprint analysis using matrix-associated laser desorption/ionization mass spectrometry, and Western blotting. Western blotting assay also revealed that detectable expression of Abeta proteins were not seen in the piglet brains. These findings indicate that TPI with sequence homology to Abeta peptides accumulates in perivascular cells of the microglia/macrophage lineage located around arterial vessels of the newborn piglet hippocampus.     

Towards understanding the mechanism of action of the multidrug resistance-linked half-ABC transporter ABCG2: a molecular modeling study

The ATP-binding cassette protein ABCG2 is a member of a broad family of ABC transporters with potential clinical importance as a mediator of multidrug resistance. We carried out a homology and knowledge-based, and mutationally improved molecular modeling study to establish a much needed structural framework for the protein, which could serve as guidance for further genetic, biochemical, and structural analyses. Based on homology with known structures of both full-length and nucleotide-binding domains (NBD) of ABC transporters and structural knowledge of integral membrane proteins, an initial model of ABCG2 was established. Subsequent refinement to conform to the lipophilic index distributions in the transmembrane domain (TMD) and to the results of site-directed mutagenesis experiments led to an improved model. The complete ABCG2 model consists of two identical subunits facing each other in a closed conformation. The dimeric interface in the nucleotide-binding domain (NBD) involves a characteristic nucleotide sandwich and the interface in the TMD consists of the TM helices 1–3 of one subunit and the helices 5 and 6 of the other. The interface between the NBD and the TMD is bridged by the conserved structural motif between TM2 and TM3, the intracellular domain 1 (ICD1), and the terminal β-strand (S6) of the central β-sheet in the NBD. The apparent flexibility of the ICD1 may play a role in transmitting conformational changes from the NBD to the TMD or from the TMD to the NBD.     

Generating Instances for MAX2SAT with Optimal Solutions

A test instance generator (an instance generator for short) for MAX2SAT is a procedure that produces, given a number n of variables, a 2-CNF formula F of n variables (randomly chosen from some reasonably large domain), and simultaneously provides one of the optimal solutions for F. We propose an outline to design an instance generator using an expanding graph of a certain type, called here an "exact 1/2-enlarger". We first show a simple algorithm for constructing an exact 1/2-enlarger, thereby deriving one concrete polynomial-time instance generator GEN. We also show that an exact 1/2-enlarger can be obtained with high probability from graphs randomly constructed. From this fact, we propose another type of instance generator RGEN; it produces a 2-CNF formula with a solution which is optimal for the formula with high probability. However, RGEN produces less structured formulas and a much larger class of formulas than GEN. In fact, we prove the NP-hardness of MAX2SAT over the set of 2-CNF formulas produced by RGEN.     

A Dynamic Motion Control Technique for Human-like Articulated Figures

This paper presents a dynamic motion control technique for human-like articulated figures in a physically based character animation system. This method controls a figure such that the figure tracks input motion specified by a user. When environmental physical input such as an external force or a collision impulse are applied to the figure, this method generates dynamically changing motion in response to the physical input. We have introduced comfort and balance control to compute the angular acceleration of the figure's joints. Our algorithm controls the several parts of a human-like articulated figure separetely through the minimum number of degrees-of-freedom. Using this approach, our algorithm simulates realistic human motions at efficient computational cost. Unlike existing dynamic simulation systems, our method assumes that input motion is already realistic, and is aimed at dynamically changing the input motion in real-time only when unexpected physical input is applied to the figure. As such, our method works efficiently in the framework of current computer games.     

Strength Analysis of Yttria-Stabilized Tetragonal Zirconia Polycrystals

Tensile strength of Y₂O₃-stabilized ZrO₂ polycrystals (Y-TZP) was measured by a newly developed tensile testing method with a rectangular bar. The tensile strength of Y-TZP was lower than that of the three-point bend strength, and the shape of the tensile strength distribution was quite different from that of the three-point bend strength distribution. It was difficult to predict the distribution curve of the tensile strength using the data of the three-point bend strength by one-modal Weibull distribution. The distribution of the tensile strength was analyzed by two- or three-model Weibull distribution coupled with an analysis of fracture origins. The distribution curve of the three-point bend strength which was estimated by multimodal Weibull distribution agreed favorably with that of the measured three-point bend strength values. A two-modal Weibull distribution function was formulated approximately from the distributions of the tensile and three-point bend strengths, and the estimated two-modal Weibull distribution function for the four-point bend strength agreed well with the measured four-point bend strength.     

Agglomeration control of hydroxyapatite nano-crystals grown in phase-separated microenvironments

Materials synthesis processes that require high temperatures consume large quantities of energy that generate an environmental burden. We attempted to synthesize hydroxyapatite (HAp) nano-crystals without firing or melting. “Water in oil” (W/O) emulsions were employed as microreactors for HAp formation. The surfactant-bounded water mediated HAp crystal nucleation, and HAp nano-crystallites were obtained. The obtained particles were aggregates composed of plate-like nano-crystals and monodisperse tiny crystals. Utilization of the W/O emulsions resulted in tunable nucleation frequency and the reactant provision, and yielded HAp nano-crystals with characteristic agglomeration properties.     

Integrable multimode interference distributed Bragg reflector laser all-optical flip-flops

A novel fully integrable all-optical flip-flop has been created using distributed Bragg reflector multimode interference bistable laser diodes. The single metal-organic vapor phase epitaxy regrowth offset quantum well structure active/passive integrated flip-flop lases in single mode at 1554 nm and is compatible with standard fabrication methods of InP photonic integrated circuits (PICs). All-optical reset switching through cross-gain saturation was demonstrated over a 52-nm range, at 1522-1574 nm. An optical set has been achieved with -3 dBm and optical reset with less than -5-dBm external light injection. The flip-flop will be useful for integration in PICs of future photonic systems for self-routing and optical memories or buffers.