このページのリンク

New Trends in Discrete and Computational Geometry / edited by János Pach
(Algorithms and Combinatorics ; 10)

データ種別 電子ブック
出版者 Berlin, Heidelberg : Springer Berlin Heidelberg
出版年 1993
本文言語 英語
大きさ XI, 340 p. 9 illus : online resource

所蔵情報を非表示

URL 電子ブック


EB0094485

書誌詳細を非表示

内容注記 I. Combinatorics and Algorithms of Arrangements
1. Introduction
2. Arrangements of Curves in the Plane
3. Lower Envelopes and Davenport-Schinzel Sequences
4. Faces in Arrangements
5. Arrangements in Higher Dimensions
6. Summary
References
II. Backwards Analysis of Randomized Geometric Algorithms
1. Introduction
2. Delaunay Triangulations of Convex Polygons
3. Intersecting Line Segments
4. Constructing Planar Convex Hulls
5. Backwards Analysis of QUICKSORT
6. A Bad Example
7. Linear Programming for Small Dimension
8. Welzl’s Minidisk Algorithm
9. Clarkson’s Backwards Analysis of the Conflict Graph Based on the Convex Hull Algorithm
10. Odds and Ends
References
III. Epsilon-Nets and Computational Geometry
1. Range Spaces and ?-Nets
2. Geometric Range Spaces
3. A Sample of Applications
4. Removing Logarithms
5. Removing the Randomization
References
IV. Complexity of Polytope Volume Computation
1. Jumps of the Derivatives
2. Exact Volume Computation is Hard
3. Volume Approximation
References
V. Allowable Sequences and Order Types in Discrete and Computational Geometry
1. Introduction
2. Combinatorial Types of Configurations in the Plane and Allowable Sequences
3. Arrangements of Lines and Pseudolines
4. Applications of Allowable Sequences
5. Order Types of Points in Rd and “Geometric Sorting”
6. The Number of Order Types in Rd
7. Isotopy and Realizability Questions
8. Lattice Realization of Order Types and the Problem of Robustness in Computational Geometry
References
VI. Hyperplane Approximation and Related Topics
1. Introduction
2. MINSUM Problem: Orthogonal L1-Fit
3. MINSUM Problem: Vertical L1-Fit
4. MINMAX Problem: Orthogonal L?-Fit
5. MINMAX Problem: Vertical L?-Fit
6. Related Issues
References
VII. Geometric Transversal Theory
1. Introduction
2. Hadwiger-Type Theorems
3. The Combinatorial Complexity of the Space of Transversals
4. Translates of a Convex Set
5. Transversal Algorithms
6. Other Directions
References
VIII. Hadwiger-Levi’s Covering Problem Revisited
0. Introduction
1. On I0(K) and I?(K)
2. On Il(K) and k-fold Illumination
3. Some Simple Remarks on H(B)
4. On Convex Bodies with Finitely Many Corner Points
5. Solution of Hadwiger-Levi’s Covering Problem for Convex Polyhedra with Affine Symmetry
References
IX. Geometric and Combinatorial Applications of Borsuk’s Theorem
1. Introduction
2. Van Kampen-Flores Type Results
3. The Ham-Sandwich Theorem
4. Centrally Symmetric Polytopes
5. Kneser’s Conjecture
6. Sphere Coverings
References
X. Recent Results in the Theory of Packing and Covering
1. Introduction
2. Preliminaries and Basic Concepts
3. A Review of Some Classical Results in the Plane
4. Economical Packing in and Covering of the Plane
5. Multiple Packing and Covering
6. Some Computational Aspects of Packing and Covering
7. Restrictions on the Number of Neighbors in a Packing
8. Selected Topics in 3 Dimensions
References
XI. Recent Developments in Combinatorial Geometry
1. The Distribution of Distances
2. Graph Dimensions
3. Geometric Graphs
4. Arrangements of Lines in Space
References
XII. Set Theoretic Constructions in Euclidean Spaces
0. Introduction
1. Simple Transfinite Constructions
2. Closed Sets or Better Well-Orderings
3. Extending the Coloring More Carefully
4. The Use of the Continuum Hypothesis
5. The Infinite Dimensional Case
6. Large Paradoxical Sets in Another Sense
References
Author Index
一般注記 Discrete and computational geometry are two fields which in recent years have benefitted from the interaction between mathematics and computer science. The results are applicable in areas such as motion planning, robotics, scene analysis, and computer aided design. The book consists of twelve chapters summarizing the most recent results and methods in discrete and computational geometry. All authors are well-known experts in these fields. They give concise and self-contained surveys of the most efficient combinatorical, probabilistic and topological methods that can be used to design effective geometric algorithms for the applications mentioned above. Most of the methods and results discussed in the book have not appeared in any previously published monograph. In particular, this book contains the first systematic treatment of epsilon-nets, geometric tranversal theory, partitions of Euclidean spaces and a general method for the analysis of randomized geometric algorithms. Apart from mathematicians working in discrete and computational geometry this book will also be of great use to computer scientists and engineers, who would like to learn about the most recent results
著者標目 Pach, János editor
SpringerLink (Online service)
件 名 LCSH:Mathematics
LCSH:Chemometrics
LCSH:Geometry
LCSH:Combinatorics
LCSH:Computational intelligence
LCSH:Economic theory
FREE:Mathematics
FREE:Combinatorics
FREE:Geometry
FREE:Economic Theory/Quantitative Economics/Mathematical Methods
FREE:Math. Applications in Chemistry
FREE:Computational Intelligence
分 類 DC23:511.6
巻冊次 ISBN:9783642580437 REFWLINK
ISBN 9783642580437
URL http://dx.doi.org/10.1007/978-3-642-58043-7
目次/あらすじ

 類似資料

 この資料を見た人はこんな資料も見ています