A new learning method using prior information of neural networks

In this paper, we present a new learning method using prior information for three-layered neural networks. Usually when neural networks are used for identification of systems, all of their weights are trained independently, without considering their interrelation of weight values. Thus the training results are not usually good. The reason for this is that each parameter has its influence on others during the learning. To overcome this problem, first, we give an exact mathematical equation that describes the relation between weight values given by a set of data conveying prior information. Then we present a new learning method that trains a part of the weights and calculates the others by using these exact mathematical equations. In almost all cases, this method keeps prior information given by a mathematical structure exactly during the learning. In addition, a learning method using prior information expressed by inequality is also presented. In any case, the degree of freedom of networks (the number of     

Performance Evaluation of Range Queries in Key Value Stores

Recently there has been a considerable increase in the number of different Key-Value stores, for supporting data storage and applications on the cloud environment. While all these solutions try to offer highly available and scalable services on the cloud, they are significantly different with each other in terms of the architecture and types of the applications, they try to support. Considering three widely-used such systems: Cassandra, HBase and Voldemort; in this paper we compare them in terms of their support for different types of query workloads. We are mainly focused on the range queries. Unlike HBase and Cassandra that have built-in support for range queries, Voldemort does not support this type of queries via its available API. For this matter, practical techniques are presented on top of Voldemort to support range queries. Our performance evaluation is based on mixed query workloads, in the sense that they contain a combination of short and long range queries, beside other types of typical queries on key-value stores such as lookup and update. We show that there are trade-offs in the performance of the selected system and scheme, and the types of the query workloads that can be processed efficiently.     

Controlled counter-flow motion of magnetic bead chains rolling along microchannels

We demonstrate controlled transport of superparamagnetic beads in the opposite direction of a laminar flow. A permanent magnet assembles 200 nm magnetic particles into about 200 μm long bead chains that are aligned in parallel to the magnetic field lines. Due to a magnetic field gradient, the bead chains are attracted towards the wall of a microfluidic channel. A rotation of the permanent magnet results in a rotation of the bead chains in the opposite direction to the magnet. Due to friction on the surface, the bead chains roll along the channel wall, even in counter-flow direction, up to at a maximum counter-flow velocity of 8 mm s⁻¹. Based on this approach, magnetic beads can be accurately manoeuvred within microfluidic channels. This counter-flow motion can be efficiently be used in Lab-on-a-Chip systems, e.g. for implementing washing steps in DNA purification.     

Cyclic moving average control approach to cylinder pressure and its experimental validation

Cyclic variability is a factor adversely affecting engine performance. In this paper a cyclic moving average regulation approach to cylinder pressure at top dead center （TDC） is proposed, where the ignition time is adopted as the control input. The dynamics from ignition time to the moving average index is described by ARMA model. With this model, a one-step ahead prediction-based minimum variance controller （MVC） is developed for regulation. The performance of the proposed controller is illustrated by experiments with a commercial car engine and experimental results show that the controller has a reliable effect on index regulation when the engine works under different fuel injection strategies, load changing and throttle opening disturbance.     

Chiral recognition ability of oligopeptide derivatives consisting of glutamyl residues

The effect of the constituting amino acid residue [Glu(OBzl)] number on the chiral recognition ability was investigated. Chiral recognition sites were prepared from oligopeptide derivatives (constituting amino acid residue number = three–five) by adopting alternative molecular imprinting. It was made clear that with a constituting amino acid residue number of four, the tetrapeptide derivative of Glu(OBzl) is the best candidate material to generate a chiral recognition site among eight types of oligopeptide derivatives in the study. The affinity constant between Ac‐L ‐Trp and a chiral recognition site ranged from 3.4 × 10³ to 1.08 × 10⁴ mol^?1 dm³, depending on the number of Glu(OBzl) residues in an oligopeptide derivative. © 2005 Wiley Periodicals, Inc. J Appl Polym Sci 95: 1302–1309, 2005     

Hemarthrosis of the knee and bone contusion

The displacement of porcine [125I] insulin bound to rat and lamprey isolated hepatocytes with unlabeled lamprey and porcine insulins was investigated. Binding affinity of lamprey insulin for insulin receptor of rat was similar to that of porcine insulin. In contrast, the binding affinity of lamprey insulin for its own insulin receptor was higher than for a rat receptor. To determine the binding affinity constants of lamprey insulin receptor, the competition binding experiments were carried out on isolated lamprey hepatocytes using lamprey insulin as unlabeled ligand and tracer. The affinity of the same binding sites on lamprey hepatocytes was assessed in similar experiments but employing porcine insulin as unlabeled ligand and tracer. It was found that while Kd of low affinity binding sites on lamprey hepatocytes were similar for lamprey and porcine insulins, the Kd of high affinity binding sites was different: the displacement curve for lamprey insulin being shifted to the left as compared to the curve for porcine insulin. The number of high and low affinity binding sites, calculated independently in Scatchard plots, was equal. We conclude that the high affinity insulin binding sites of lamprey but not of rat hepatocytes reveal some species specificity in ligand-receptor interaction.     

Seismic Earth Pressure Exerted on Retaining Walls Under a Large Seismic Load

In recent years, serious damage has been done to retaining structures because of large earthquakes. In order to establish practical methods for evaluating the seismic earth pressure, which is one of the important external forces acting on retaining structures during large earthquakes, a series of shaking table tests was conducted on retaining wall (RW) models. The experiments revealed that the seismic active earth pressure was considerably smaller than that obtained by the Mononobe-Okabe theory, particularly under a large seismic load. Furthermore, it was demonstrated that the seismic earth pressure had an upper limit, which was determined by the force equilibrium of the soil wedge at the critical state when the RW lost its stability. On the basis of the test results, a new method to evaluate the seismic earth pressure for practical designs under a large seismic load has been suggested. This proposed method provides a reasonable earth pressure as well as an angle of failure plane, those of which depend on the seismic stability of the retaining wall. It has been confirmed that earth pressure obtained by the proposed method agrees well with the measured seismic earth pressure exerted on several retaining walls with different degrees of stability.     

Multiple modulations of action potential duration by different calcium channel blocking agents in guinea pig ventricular myocytes

Morphometric and functional brain research is now an integral part of the most concentrated effort in a century to understand some of the most distressing illnesses to which man can fall victim. It will undoubtedly expand. The only ones with anything to fear from this work are those who, for ideological reasons, cannot accept its basic premises. The ethical issues it raises are not in general insurmountable, though the questions posed by the concept of informed consent remain.     

Cloning, expression, and characterization of a cDNA encoding a novel human growth factor for primitive hematopoietic progenitor cells

Multiple growth factors synergistically stimulate proliferation of primitive hematopoietic progenitor cells. A human myeloid cell line, KPB-M15, constitutively produces a novel hematopoietic cytokine, termed stem cell growth factor (SCGF), possessing species-specific proliferative activities. Here we report the molecular cloning, expression, and characterization of a cDNA encoding human SCGF using a newly developed lambdaSHDM vector that is more efficient for differential and expression cloning. cDNA for SCGF encodes a 29-kDa polypeptide without N-linked glycosylation. SCGF transiently produced by COS-1 cells supports growth of hematopoietic progenitor cells through a short-term liquid culture of bone marrow cells and exhibits promoting activities on erythroid and granulocyte/macrophage progenitor cells in primary semisolid culture with erythropoietin and granulocyte/macrophage colony-stimulating factor, respectively. Expression of SCGF mRNA is restricted to myeloid cells and fibroblasts, suggesting that SCGF is a growth factor functioning within the hematopoietic microenvironment. SCGF could disclose some human-specific mechanisms as yet unidentified from studies on the murine hematopoietic system.