Black Box Low Tensor-Rank Approximation Using Fiber-Crosses

In this article we introduce a black box type algorithm for the approximation of tensors A in high dimension d. The algorithm adaptively determines the positions of entries of the tensor that have to be computed or read, and using these (few) entries it constructs a low rank tensor approximation X that minimizes the ? ₂-distance between A and X at the chosen positions. The full tensor A is not required, only the evaluation of A at a few positions. The minimization problem is solved by Newton’s method, which requires the computation and evaluation of the Hessian. For efficiency reasons the positions are located on fiber-crosses of the tensor so that the Hessian can be assembled and evaluated in a data-sparse form requiring a complexity of $\mathcal{O}(Pd)$ , where P is the number of fiber-crosses and d the order of the tensor.     

Peeking Inside the Black Box: Visualizing Statistical Learning With Plots of Individual Conditional Expectation

This article presents individual conditional expectation (ICE) plots, a tool for visualizing the model estimated by any supervised learning algorithm. Classical partial dependence plots (PDPs) help visualize the average partial relationship between the predicted response and one or more features. In the presence of substantial interaction effects, the partial response relationship can be heterogeneous. Thus, an average curve, such as the PDP, can obfuscate the complexity of the modeled relationship. Accordingly, ICE plots refine the PDP by graphing the functional relationship between the predicted response and the feature for individual observations. Specifically, ICE plots highlight the variation in the fitted values across the range of a covariate, suggesting where and to what extent heterogeneities might exist. In addition to providing a plotting suite for exploratory analysis, we include a visual test for additive structure in the data-generating model. Through simulated examples and real datasets, we demonstrate how ICE plots can shed light on estimated models in ways PDPs cannot. Procedures outlined are available in the R package ICEbox.     

Criteria for Componentwise Linearity

We establish characteristic-free criteria for the componentwise linearity of graded ideals. As applications, we classify the componentwise linear ideals among the Gorenstein ideals, the standard determinantal ideals, and the ideals generated by the submaximal minors of a symmetric matrix.     

Linearity measure for curve segments

In this paper, we consider linearity measure for a bounded length curves. First, we define a new linearity measure for open curve segments, and then extend method to closed curves (contours). The derived measures (for both, open curve segments and closed curves) are invariant with respect to similarity transformations. The linearity measure for open curve segments picks the value 1 if and only if the measured open line segment is a perfect straight line segment while the established linearity measures for closed curves never reach 1, as preferred.     

论“打开黑箱评价”的网络DEA模型

讨论了两阶段网络DEA模型KH,复合网络DEA模型WYP,以及两个模型之间的关系.并指出了上述两个模型能够打开"黑箱".     

Linearity of the core correspondence

Bloch and de Clippel (J Econ Theory 145:2424–2434, 2010) characterized sets of balanced TU-games on which the core correspondence is linear by means of an equivalence relation. We characterize maximal regions on which the core correspondence is linear in four different ways. First, by finitely many linear equalities and inequalities; thus, the core is piecewise linear. Second, maximal linear regions coincide with closures of equivalence classes (in the sense of Bloch and de Clippel) that are maximal w.r.t. set inclusion. Third, maximal linear regions coincide with closures of equivalence classes of full dimension. Fourth, for every extreme point of the core of a game in the interior of a maximal linear region, the collection of tight core inequalities constitutes a basis.     

Linearity criterion for the selection of an optimal tree

The aim of this paper is to present a method for selecting the optimal tree among the possible trees that can be generated starting from the a data set. Analysis a quantity criterion is used through the linear combination of the quality measurements of the tree, namely, resubstitution error and linearity. The application of the method leads to a succession of optimal trees, in such a way, that an element of the succession is associated with each possible value of the linear combination's parameter α.     

Forcing Linearity Numbers for Multiplication Modules

In this article, we prove that for any multiplication module M, the forcing linearity number of M, fln(M), belongs to {0,1,2}, and if M is finitely generated whose annihilator is contained in only finitely many maximal ideals, then fln(M) = 0. Also, the forcing linearity numbers of multiplication modules over some special rings are given. We also show that every multiplication module is semi-endomorphal.     

Constrained Global Optimization of Expensive Black Box Functions Using Radial Basis Functions

We present a new strategy for the constrained global optimization of expensive black box functions using response surface models. A response surface model is simply a multivariate approximation of a continuous black box function which is used as a surrogate model for optimization in situations where function evaluations are computationally expensive. Prior global optimization methods that utilize response surface models were limited to box-constrained problems, but the new method can easily incorporate general nonlinear constraints. In the proposed method, which we refer to as the Constrained Optimization using Response Surfaces (CORS) Method, the next point for costly function evaluation is chosen to be the one that minimizes the current response surface model subject to the given constraints and to additional constraints that the point be of some distance from previously evaluated points. The distance requirement is allowed to cycle, starting from a high value (global search) and ending with a low value (local search). The purpose of the constraint is to drive the method towards unexplored regions of the domain and to prevent the premature convergence of the method to some point which may not even be a local minimizer of the black box function. The new method can be shown to converge to the global minimizer of any continuous function on a compact set regardless of the response surface model that is used. Finally, we considered two particular implementations of the CORS method which utilize a radial basis function model (CORS-RBF) and applied it on the box-constrained Dixon–Szegö test functions and on a simple nonlinearly constrained test function. The results indicate that the CORS-RBF algorithms are competitive with existing global optimization algorithms for costly functions on the box-constrained test problems. The results also show that the CORS-RBF algorithms are better than other algorithms for constrained global optimization on the nonlinearly constrained test problem.     

Organizations and Complexity: Searching for the Edge of Chaos

Traditional organizational theory advocates increased differentiation and horizontal integration for organizations in unstable environments or with uncertain technologies. This paper seeks to develop a better understanding of the relationship of group structure and the level of interdependency between individuals on group performance under various task complexities. Complexity theory in general, and NK models in particular, are introduced as theoretical frameworks that offer an explanation for group performance. Simulation models are developed, based on the communication network research of Bavelas (1948) and Leavitt (1952), to explore the effects of decentralization and interdependence. The simulation model developed here shows general consistency with previous human subject experiments. However, contrary to predictions, not all decentralized group structures perform well when undertaking complex task assignments. Structures that are highly connected (actors communicating with all others) perform much worse than those with a lower level of connection. Further experiments varying both the number of actors and the degree of interdependence between them find evidence of the edge of chaos. This research advances our understanding of organizations beyond earlier models by suggesting that there is an optimal range of interconnectedness between actors or tasks that explains the variation in performance. An intriguing result is that this optimal level of interdependence is fairly low, regardless of the size of the group.     

Middle Grades: Misconceptions in Statistical Thinking

A sample of 134 sixth‐grade students who were using the Connected Mathematics curriculum were administered an open‐ended item entitled, Vet Club (Balanced Assessment, 2000). This paper explores the role of misconceptions and naïve conceptions in the acquisition of statistical thinking for middle grades students. Students exhibited misconceptions and naïve conceptions regarding representing data graphically, interpreting the meaning of typicality, and plotting 0 above the x‐axis.     

Searching for losers

We investigate the following process: N people select b losers by flipping coins. The 0-party continues until there are less than b losers; then the 1-party has to find the other losers by the same process. The average time for this process is about long₂ N, but this result requires rather advanced methods. Furthermore, the average size of a binary tree associated to this process and the average number of coin flippings are computed. The method used in this article can be used to give asympotical solutions of a special type of bivariate recurrences. © 1993 John Wiley & Sons, Inc.     

Forcing Linearity Numbers for Modules over PID's

Let V be a module over a principal ideal domain. Then V = M N where M is divisible and N has no nonzero divisible submodules. In this paper we determine the forcing linearity number for V when N is a direct sum of cyclic modules. As a consequence, the forcing linearity numbers of several classes of Abelian groups are obtained.     

Polytropes and Tropical Eigenspaces: Cones of Linearity

The map which takes a square matrix A to its polytrope is piecewise linear. We show that cones of linearity of this map form a polytopal fan partition of $\mathbb{R}^{n \times n}$ , whose face lattice is anti-isomorphic to the lattice of complete set of connected relations. This fan refines the non-fan partition of $\mathbb{R}^{n \times n}$ corresponding to cones of linearity of the eigenvector map. Our results answer open questions in a previous work with Sturmfels (Bull. Lond. Math. Soc. 54:27–36, 2013) and lead to a new combinatorial classification of polytropes and tropical eigenspaces.