In this paper we propose linear-time CUR approximation algorithms for admissible matrices obtained from the hierarchical form of Boundary Element matrices. We propose a new approach called geometric sampling to obtain...
详细信息
In this paper we propose linear-time CUR approximation algorithms for admissible matrices obtained from the hierarchical form of Boundary Element matrices. We propose a new approach called geometric sampling to obtain indices of most significant rows and columns using information from the domains where the problem is posed. Our strategy is tailored to Boundary Element Methods (BEM) since it uses directly and explicitly the cluster tree containing information from the problem geometry. Our CUR algorithm has precision comparable with low-rank approximations created with the truncated QR factorization with column pivoting (QRCP) and the Adaptive Cross Approximation (ACA) with full pivoting, which are quadratic-cost methods. When compared to the well-known linear-time algorithm ACA with partial pivoting, we show that our algorithm improves, in general, the convergence error and overcomes some cases where ACA fails. We provide a general relative error bound for CUR approximations created with geometrical sampling. Finally, we evaluate the performance of our algorithms on traditional BEM problems defined over different geometries. (C) 2019 Elsevier B.V. All rights reserved.
We establish decay bounds for the entries of f(A), where A is a sparse (in particular, banded) n x n diagonalizable matrix and f is smooth on a subset of the complex plane containing the spectrum of A. Combined with t...
详细信息
We establish decay bounds for the entries of f(A), where A is a sparse (in particular, banded) n x n diagonalizable matrix and f is smooth on a subset of the complex plane containing the spectrum of A. Combined with techniques from approximation theory, the bounds are used to compute sparse (or banded) approximations to f(A), resulting in algorithms that under appropriate conditions have linear complexity in the matrix dimension. Applications to various types of problems are discussed and illustrated by numerical examples.
The class of bipartite permutation graphs is the intersection of two well known graph classes: bipartite graphs and permutation graphs. A complete bipartite decomposition of a bipartite permutation graph is proposed i...
详细信息
The class of bipartite permutation graphs is the intersection of two well known graph classes: bipartite graphs and permutation graphs. A complete bipartite decomposition of a bipartite permutation graph is proposed in this note. The decomposition gives a linear structure of bipartite permutation graphs, and it can be obtained in O(n) time, where n is the number of vertices. As an application of the decomposition, we show an O(n) time and space algorithm for finding a longest path in a bipartite permutation graph. (C) 2007 Elsevier B.V. All rights reserved.
Reasoning with minimal models is at the heart of many knowledge-representation systems. Yet it turns out that this task is formidable, even when very simple theories are considered. In this paper, we introduce the eli...
详细信息
Reasoning with minimal models is at the heart of many knowledge-representation systems. Yet it turns out that this task is formidable, even when very simple theories are considered. In this paper, we introduce the elimination algorithm, which performs, in lineartime, minimal model finding and minimal model checking for a significant subclass of positive CNF theories which we call positive bead-cycle-free (HCF) theories. We also prove that the task of minimal entailment is easier for positive HCF theories than it is for the class of all positive CNF theories. Finally, we show how variations of the elimination algorithm can be applied to allow queries posed on disjunctive deductive databases and disjunctive default theories to be answered in an efficient way. (C) 1997 Published by Elsevier Science B.V.
In scheduling theory it is widely assumed that a task is to be processed on one processor at a time. This assumption is not so obvious in the context of recently emerging parallel computer systems and parallel algorit...
详细信息
In scheduling theory it is widely assumed that a task is to be processed on one processor at a time. This assumption is not so obvious in the context of recently emerging parallel computer systems and parallel algorithms. In this work we consider tasks requiring more than one dedicated processor at a time, i.e. sets of processors simultaneously. linear time algorithms will be given for the case of two, three and four processors and the L(max) criterion. The algorithms are based on the same simple paradigm. In some cases they deliver optimal solutions. In other cases, optimality is not guaranteed but they can still be used as fast approximation algorithms for which the worst case performance bounds are given. Results of the computational experiments involving four processors are reported.
The isotonic regression problem has applications in statistics, operations research, and image processing. In this paper a general framework for the isotonic regression algorithm is proposed. Under this framework, we ...
详细信息
The isotonic regression problem has applications in statistics, operations research, and image processing. In this paper a general framework for the isotonic regression algorithm is proposed. Under this framework, we discuss the isotonic regression problem in the case where the directed graph specifying the order restriction is a directed tree with n vertices. A new algorithm is presented for this case, which can be regarded as a generalization of the PAV algorithm of Ayer et al. Using a simple tree structure such as the binomial heap, the algorithm can be implemented in O(n log n) time, improving the previously best known O(n(2)) time algorithm. We also present linear time algorithms for special cases where the directed graph is a path or a star.
Approximation algorithms for embedding hyperedges in a cycle so as to minimize the maximum congestion are presented. Our algorithms generate an embedding by transforming the problem into another problem solvable in po...
详细信息
Approximation algorithms for embedding hyperedges in a cycle so as to minimize the maximum congestion are presented. Our algorithms generate an embedding by transforming the problem into another problem solvable in polynomial time. One algorithm transforms it to a linear programming problem, and the other one to the problem of embedding edges in a cycle. Both algorithms generate an embedding with congestion at most twice of that in an optimal solution. Our problem has applications in CAD and parallel computation. (C) 1998 Elsevier Science B.V. All rights reserved.
The definition of similarity measures for phylogenetic trees has been motivated by the computation of consensus trees, the search by similarity in databases, and the assessment of phylogenetic reconstruction methods. ...
详细信息
The definition of similarity measures for phylogenetic trees has been motivated by the computation of consensus trees, the search by similarity in databases, and the assessment of phylogenetic reconstruction methods. The transposition distance for fully resolved trees is a recent addition to the extensive collection of available metrics for comparing phylogenetic trees. In this work, we generalize the transposition metric from fully resolved to arbitrary phylogenetic trees, through a construction that involves an embedding of the set of phylogenetic trees (up to isomorphisms) with a fixed number of labeled leaves into a symmetric group. We also show that this transposition distance can be computed in lineartime and we establish some of its basic properties. (C) 2009 Elsevier Ltd. All rights reserved.
The problem of fitting n data points by an integer quasi-convex (also quasi-concave, umbrella or unimodal) function using the weighted least squares distance function is considered. An algorithm of lineartime (O(n)) ...
详细信息
The problem of fitting n data points by an integer quasi-convex (also quasi-concave, umbrella or unimodal) function using the weighted least squares distance function is considered. An algorithm of lineartime (O(n)) worst-case complexity and thus optimal is constructed for computing a best fit. This problem arises in the context of curve fitting or statistical estimation. (C) 2009 Elsevier Ltd. All rights reserved.
This paper concerns with computation of topological invariants such as genus and the Betti numbers. We design a lineartime algorithm that determines such invariants for digital spaces in 3D. These computations could ...
详细信息
This paper concerns with computation of topological invariants such as genus and the Betti numbers. We design a lineartime algorithm that determines such invariants for digital spaces in 3D. These computations could have applications in medical imaging as they can be used to identify patterns in 3D image. Our method is based on cubical images with direct adjacency, also called (6, 26)-connectivity images in discrete geometry. There are only six types of local surface points in such a digital surface. Two mathematical ingredients are used. First, we use the Gauss-Bonnett Theorem in differential geometry to determine the genus of 2-dimensional digital surfaces. This is done by counting the contribution for each of the six types of local surface points. The new formula derived in this paper that calculates genus is g = 1 + (vertical bar M-5 vertical bar + 2 . vertical bar M-6 vertical bar - vertical bar M-3 vertical bar)/8 where M-i indicates the set of surface-points each of which has i adjacent points on the surface. Second, we apply the Alexander duality to express the homology groups of a 3D manifold in the usual 3D space in terms of the homology groups of its boundary surface. While our result is stated for digital spaces, the same idea can be applied to simplicial complexes in 3D or more general cell complexes. (C) 2010 Elsevier B.V. All rights reserved.
暂无评论