ERM learning with unbounded sampling

The learning approach of empirical risk minimization (ERM) is taken for the regression problem in the least square framework. A standard assumption for the error analysis in the literature is the uniform boundedness of the output sampling process. In this paper we abandon this boundedness assumption and conduct error analysis for the ERM learning algorithm with unbounded sampling processes satisfying an increment condition for the moments of the output. The key novelty of our analysis is a covering number argument for estimating the sample error.     

Multi‐layered Youden Rectangles

Multi‐layered Youden rectangles are introduced. These new designs include double and triple Youden rectangles as subdesigns. Though many double Youden rectangles are known, the triples, introduced in 1994, have to now yielded few nontrivial examples. Two infinite series of multi‐layered Youden rectangles are constructed, and so also many new triple Youden rectangles.     

Large sets of Hamilton cycle and path decompositions

The existence spectrums for large sets of Hamilton cycle decompositions and Hamilton path decompositions are completed. Also, we show that the completion of large sets of directed Hamilton cycle decompositions and directed Hamilton path decompositions depends on the existence of certain special tuscan squares. Several conjectures about special tuscan squares are posed.     

Topological estimation of proton-ligand formation constants of potential antitumour agents: Salicylhydroxamic acids

Proton-ligand formation constants of salicylhydroxamic acids (SHA) and their nuclear substituted derivatives have been estimated topologically using the normalized Wiener index, referred to as mean square Wiener index (Wms). Regression analysis of the data indicates that Wms can be used successfully for estimating and monitoring proton-ligand formation constants.     

Optimal doubly constant weight codes

A doubly constant weight code is a binary code of length n₁ + n₂, with constant weight w₁ + w₂, such that the weight of a codeword in the first n₁ coordinates is w₁. Such codes have applications in obtaining bounds on the sizes of constant weight codes with given minimum distance. Lower and upper bounds on the sizes of such codes are derived. In particular, we show tight connections between optimal codes and some known designs such as Howell designs, Kirkman squares, orthogonal arrays, Steiner systems, and large sets of Steiner systems. These optimal codes are natural generalization of Steiner systems and they are also called doubly Steiner systems. © 2007 Wiley Periodicals, Inc. J Combin Designs 16: 137–151, 2008     

Estimates of a certain sum involving coefficients of cusp forms in weight and level aspects

We consider the sum of coefficients which are in the Dirichlet series expansion of symmetric square L-functions. In this paper, we obtain two estimates of this sum in weight and level aspects. These imply two estimates of the sum of the n ²th Fourier coefficients of cusp forms.     

Analysis of ordered categorical data from repeated measurements assuming a quantitative latent variable

Summary The purpose of the present paper is to propose an analytical method for ordered categorical responses obtained from a repeated measurement/longitudinal experiment. The ordered categorical scale is assumed to be a manifestation of a latent quantitative variable. A linear model is assumed for location parameters of the underlying distributions. Weighted least square method is applied to parameter estimation and subsequent analysis. Two data sets are analyzed to show several aspects of analysis by the proposed model and to discuss comparative characteristics of analysis compared with earlier analysis. A mention is made for a computer software program for the proposed model.     

Adaptive nonparametric estimation of a multivariate regression function

We consider the kernel estimation of a multivariate regression function at a point. Theoretical choices of the bandwidth are possible for attaining minimum mean squared error or for local scaling, in the sense of asymptotic distribution. However, these choices are not available in practice. We follow the approach of Krieger and Pickands (Ann. Statist.9 (1981) 1066–1078) and Abramson (J. Multivariate Anal.12 (1982), 562–567) in constructing adaptive estimates after demonstrating the weak convergence of some error process. As consequences, efficient data-driven consistent estimation is feasible, and data-driven local scaling is also feasible. In the latter instance, nearest-neighbor-type estimates and variance-stabilizing estimates are obtained as special cases.     

Two‐dimensional prediction of time dependent,turbulent flow around a square cylinder confined in a channel

This paper presents two‐dimensional and unsteady RANS computations of time dependent, periodic, turbulent flow around a square block. Two turbulence models are used: the Launder–Sharma low‐Reynolds number k–ε model and a non‐linear extension sensitive to the anisotropy of turbulence. The Reynolds number based on the free stream velocity and obstacle side is Re=2.2×10⁴. The present numerical results have been obtained using a finite volume code that solves the governing equations in a vertical plane, located at the lateral mid‐point of the channel. The pressure field is obtained with the SIMPLE algorithm. A bounded version of the third‐order QUICK scheme is used for the convective terms. Comparisons of the numerical results with the experimental data indicate that a preliminary steady solution of the governing equations using the linear k–ε does not lead to correct flow field predictions in the wake region downstream of the square cylinder. Consequently, the time derivatives of dependent variables are included in the transport equations and are discretized using the second‐order Crank–Nicolson scheme. The unsteady computations using the linear and non‐linear k–ε models significantly improve the velocity field predictions. However, the linear k–ε shows a number of predictive deficiencies, even in unsteady flow computations, especially in the prediction of the turbulence field. The introduction of a non‐linear k–ε model brings the two‐dimensional unsteady predictions of the time‐averaged velocity and turbulence fields and also the predicted values of the global parameters such as the Strouhal number and the drag coefficient to close agreement with the data. Copyright © 2009 John Wiley & Sons, Ltd.     

On the range of influences in back-circulant Latin squares

In this note we obtain a large lower bound for the index of a certain critical set in the back-circulant Latin squares of odd order. This resolves in the negative a conjecture of Fitina, Seberry and Chaudhry [Back-circulant Latin square and the influence of a set, Austral. J. Combin. 20 (1999) 163-180].