A Br0nsted. An Introduction to Convex Polytopes. Graduate Texts in Mathematics. Springer-Verlag, New York, 1983.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....to the by projecting the points of S onto the paraboloid of revolution U: x 4 = i= x i. upper bound theorem for convex polytopes, the maximum numbers of 1 , 2 , and 3 faces of a convex I 2 2 I 2 polytope with n 5 vertices in 4 are at most (n n) 3n, and respectively (see for example [6,57]) The lifting map implies the same upper bounds for IFil, with 1 As a matter of fact, the upper bound for IF.3I is one less than for the number of 3 faces of conv( v) because at least one 3 face belongs to the upper boundary of conv( v) By a result of [74] these bounds are tight even though the ....

A Br0nsted. An Introduction to Convex Polytopes. Graduate Texts in Mathematics. Springer-Verlag, New York, 1983.

....The following two theorems from polyhedral theory provide the necessary tools. Although the following theorems hold for d polytopes in R , we restrict our attention to the case where d = 3. Before stating the theorems, we review some terminology. For more details, the reader is referred to [Br81], or [Gru67] Given a convex polyhedron P , we denote its open interior by int(P ) its open exterior by ext(P ) and its boundary by P . The boundary is considered part of the polyhedron, i.e. P = P [ int(P ) A face F of a polyhedron P is a vertex, an edge or a facet of P . The dimension of a ....

....that H int(P ) we say that v is beneath H, or beyond H, with respect to P ) provided v belongs to the open halfspace determined by H which contains int(P ) or does not meet P . Given a set of points S in R , CH (S ) denotes the convex hull of the points. Theorem 2. 1 (Theorem 11.11 in [Br81]) Let P be a convex polyhedron in R . Let V represent the vertices of P . Let H be a plane in R with H int(P ) 6= and H V = and let K be one of the two closed halfspaces bounded by H. Then we have: 1. The set P = K P is a convex polyhedron and H P is a facet of P . 2. ....

Arno Brndsted. Introduction to Convex Polytopes. Springer Verlag, 1981.

....00 contained in H, and vertices e 00 and f 00 corresponding to the original edges e and f . Now all we want is a path of edges from e 00 to f 00 which misses H 00 ; those edges will then correspond to the desired two dimensional faces in P . This now follows easily from Theorem 15.4 in [8]. Lemma 7 Any two paths e = fv i g; fe i g) f = fw i g; ff i g) increasing relative to the same L from v 0 = w 0 = to v s = w t = with L maximal in the polytope at may be transformed into each other in steps, with each step involving only a transformation within a single 2 dimensional ....

A. Brndsted, An Introduction to Convex Polytopes, Graduate Texts in Mathematics 90, Springer-Verlag, 1983.

....by g i the number of vertices of P such that ind (v) i. De ne g vector of a polytope P to be a sequence g = g 0 ; g 1 ; g d ) It is known that when P is simple, the g vector does not depend on , and (4) F (q) G(q 1) where G(q) P k g k q k , F (q) P k f k q k (see [2,15]) Of all di erent linear functions , one will be particularly useful for our purposes. Let E = fe 1 ; e 2 ; e m g be a linear basis in the vector space V . De ne a lexicographic orientation of edges of the polytope P by the following rule. Let edge (v 1 ; v 2 ) be oriented from v 1 to v ....

A. Brndsted, An Introduction to Convex Polytopes, Springer, Berlin, 1985.

....of Clarkson and Shor [8] and the derandomized algorithm of Chazelle [6] In some sense the algorithms [18] and [6] can be considered optimal. The upper bound theorem of McMullen states that for any polytope P defined by m halfspaces, size(P ) O(m bd=2c ) and this bound is achieved (see, e.g. [3]) The algorithms [18] and [6] solve the vertex enumeration problem in this time bound. However, it is also well known that size(P ) Omega Gamma md) and this bound is also achieved. An efficient vertex enumeration algorithm for such polytopes should clearly be polynomial in md. It is not known ....

.... d is called a convex set if every convex combination of points in K is also in K. The convex hull of X , denoted conv(X) is the set of all convex combinations of X . The relative interior of a convex set K, written relint(K) is the interior of K in the affine hull of K. Lemma 1 (Theorem 3. 3 [3]) For any convex set C, any point x in relint(C) and any point y 2 C, the open line segment from x to y is contained in relint(C) Lemma 2 (Theorem 3.5 [3] For any convex set C and any point x 2 C, the following conditions are equivalent: a) x 2 relint(C) b) For any y 2 C distinct from x, ....

[Article contains additional citation context not shown here]

A. Brøndsted. Introduction to Convex Polytopes. Springer Verlag, 1981.

.... [f1; 2g] f3; 5g] f4; 6g] f7g] f8; 9g] f10g] i and LOT = h [f1; 2g] f3; 4; 5; 6g] f7g] f8; 9; 10g] i : We denote by A 4 B the symmetric di erence of two sets A and B, that is, A 4B : A n B) B n A) Basic concepts in polyhedral theory can be found in Gr unbaum [3] or Br nsted [2]. De nitions and more general results on adjacency of vertices on 0 1 polytopes can be found in Hausmann [6] In the next lemma we mention some equivalent formulations of the adjacency concept to be used in this paper (see Pulleyblank [9] a) c) b) 1 2 7 8 9 10 1 2 3 4 7 5 6 3 4 5 6 3 4 5 6 ....

A. Brndsted, An Introduction to Convex Polytopes (Springer{Verlag, New York, 1982).

....Polytopes P and Q are combinatorially equivalent (respectively dual) if their face lattices are isomorphic (respectively anti isomorphic, i.e. isomorphic with the direction of inclusion reversed) For any point set X, the polar X of X is defined as f y j Xy 1 g. It is known (see e.g. [5]) that if P is a polytope containing the origin in its interior, P is a polytope dual to P , containing the origin in its interior. A family of polytopes is used here to mean an infinite set of polytopes. Usually, but not necessarily, families arise in some natural way from a problem such as ....

A. Brøndsted. Introduction to Convex Polytopes. Springer-Verlag, 1981.

....of D u ; that is, D u may be expressed as the convex hull of its extreme points, D u =conv[ext(D u ) 2 Proof. The proof follows from the fact that D u is a bounded set defined by a finite intersection of closed half spaces, see (26) and (27) Then by definition, D u is a polytope [13] and Theorem 3.4 follows directly from properties of polytopes [13] 2 The following theorem establishes a relationship between scheduling policies and the vertices of D u . Theorem 3.5 d is a vertex (extreme point) of the set D u iff d is a dropping rate vector resulting from a ....

....extreme points, D u =conv[ext(D u ) 2 Proof. The proof follows from the fact that D u is a bounded set defined by a finite intersection of closed half spaces, see (26) and (27) Then by definition, D u is a polytope [13] and Theorem 3. 4 follows directly from properties of polytopes [13]. 2 The following theorem establishes a relationship between scheduling policies and the vertices of D u . Theorem 3.5 d is a vertex (extreme point) of the set D u iff d is a dropping rate vector resulting from a Deadline Sensitive Ordered HoL (DSO HoL) priority service policy, ....

[Article contains additional citation context not shown here]

Arne Brøndsted. An Introduction to Convex Polytopes. Springer-Verlag, 1983.

....proper faces of P . Pick two vertices u and v of P such that ffi(P ) d P (u; v) It is well known that there exists a (possibly empty) path on the 1 skeleton of P from u to some vertex u 2 F (P; w 1 ) where the vertices along the path attain strictly increasing values under w 1 (see e.g. [1]) Similarly, there exists a strictly increasing path from v to some v 2 F (P; w 1 ) and strictly decreasing paths from u and v to some vertices u Gamma and v Gamma in F Gamma (P; w 1 ) respectively. Research supported by the German Israeli Foundation, grant no. I 84 095. 6 88 y ....

A. Brøndsted, "An Introduction to Convex Polytopes," Springer-Verlag, Berlin, New York, 1983.

....by g i the number of vertices of P such that ind (v) i. Define g vector of a polytope P to be a sequence g = g 0 ; g 1 ; g d ) It is known that when P is simple the g vector does not depend on , and (4) F (q) G(q 1) where G(q) P k g k q k , F (q) P k f k q k (see [Br,Z]) Of all different linear functions , one will be particularly useful for our purposes. Let E = fe 1 ; e 2 ; e m g be a linear basis in the vector space V . Define a lexicographic orientation of edges of the polytope P by the following rule. Let edge (v 1 ; v 2 ) be oriented from v 1 to ....

A. Brøndsted, An Introduction to Convex Polytopes, Springer, Berlin, 1985.

..... This is somewhat surprising, since the graph of P (S n ) is Hamilton connected [19] and its diameter is 2 for all n 4 [1] Before going on, we recall some definitions and properties of convex polytopes and oriented matroids. For the theory of convex polytopes, consult, for example, 10] [8], and for oriented matroids [7] A convex polytope P is k neighborly if every k subset of its vertices is a (k Gamma 1) face of P . If P is k neighborly, then it is i neighborly for i = 0; 1; Delta Delta Delta ; k. We define the neighborliness degree of P to be the largest k for which it is ....

A. Brøndsted, "An Introduction to Convex Polytopes," Springer-Verlag, Berlin, New York, 1983.

....OF OSLO Department of informatics Polytopes related to some polyhedral norms. Geir Dahl Report 226, ISBN 82 7368 140 8 November Polytopes related to some polyhedral norms. Geir Dahl # November 1996 Submitted to Linear Algebra and its Applications. Abstract Let K = K n K n [0, 1] # IR 2n 1 where K n is the standard simplex in IR n , i.e. K n = x # IR n : P n j=1 x j = 1 . We consider the set P # IR 2n 1 given by P = x, y, z) # K : #x y## # z (where #x y## = max j#n x j y j ) We call P the l # distance polytope in IR n . These polytopes are ....

....We shall study a set P related to the l # distance between vectors in the standard simplex. Let N = 1, n where n # 2 is a natural number and denote the ith unit vector in IR n by e i . The notation x(N) is used for the sum P j#N x j when x # IR n . De ne K = K n K n [0, 1] where K n is the standard # Institute of Informatics, University of Oslo, P.O.Box 1080, Blindern, 0316 Oslo, Norway (Email:geird i .uio.no) simplex in IR n , i.e. K n = x # IR n : P n j=1 x j = 1 . In addition, we de ne the simplex K # n = x # IR n : P n j=1 x j # 1 . ....

[Article contains additional citation context not shown here]

A. Br#ndsted. An introduction to convex polytopes. Springer, New York, 1983.

....that the conditions in (5) and (6) are sufficient [8] That is, the performance given by any vector in the region describe by (5) and (6) is delivered by some work conserving policy f . Let D denote the collection of all vectors d satisfying (5) and (6) Then by definition D is a convex polytope [15]. Using results from convex polytopes [15] any vector in the set D can be expressed as a convex combination of extreme points (vertices) of D; that is, D may be expressed as the convex hull of its extreme points, D=conv[exp(D) b 1 2 b b 1,2 D Dropping rate, source 1 Dropping rate, ....

....[8] That is, the performance given by any vector in the region describe by (5) and (6) is delivered by some work conserving policy f . Let D denote the collection of all vectors d satisfying (5) and (6) Then by definition D is a convex polytope [15] Using results from convex polytopes [15], any vector in the set D can be expressed as a convex combination of extreme points (vertices) of D; that is, D may be expressed as the convex hull of its extreme points, D=conv[exp(D) b 1 2 b b 1,2 D Dropping rate, source 1 Dropping rate, source 2 (1,2) 2,1) 2,1,3) 2,3,1) ....

Arne Brøndsted. An Introduction to Convex Polytopes. Springer-Verlag, 1983.

....of D u ; that is, D u may be expressed as the convex hull of its extreme points, D u =conv[ext(D u ) 2 Proof. The proof follows from the fact that D u is a bounded set defined by a finite intersection of closed half spaces, see (24) and (25) Then by definition, D u is a polytope [14] and Theorem 3.4 follows directly from properties of polytopes [14] 2 The following theorem establishes a relationship between scheduling policies and the vertices of D u . b 1 2 b b 1,2 D u Dropping rate, source 1 Dropping rate, source 2 (1,2) 2,1) 2,1,3) 2,3,1) 3,2,1) ....

....extreme points, D u =conv[ext(D u ) 2 Proof. The proof follows from the fact that D u is a bounded set defined by a finite intersection of closed half spaces, see (24) and (25) Then by definition, D u is a polytope [14] and Theorem 3. 4 follows directly from properties of polytopes [14]. 2 The following theorem establishes a relationship between scheduling policies and the vertices of D u . b 1 2 b b 1,2 D u Dropping rate, source 1 Dropping rate, source 2 (1,2) 2,1) 2,1,3) 2,3,1) 3,2,1) 3,1,2) 1,3,2) 1,2,3) Dropping rate, source 1 Dropping rate, source ....

Arne Brøndsted. An Introduction to Convex Polytopes. Springer-Verlag, 1983.

....vector of S (so # S v equals 1 if v # S and 0 otherwise) and we also de ne x(S) P v#S x v for x # IR V . If A # IR V the convex hull (conical hull) of A is denoted by conv(A) cone(A) The relative interior of a convex set C is denoted by rint(C) For polyhedral theory, we refer to [4], 12] 14] Some graph terminology is used, but is is fairly standard. 2 A cone of row ordered vectors Let n i for i = 1, m be given positive integers. Let R i = i, j) 1 # j # n i for i = 1, m and de ne the index set (or node set) V = R 1 # . #Rm . Each set R i is ....

....x # C(N ) satis es x u = 0 and, in particular, this holds for each x # C(M) as C(N ) # C(M) This proves that N # M. Let FC denote the set of all faces of the cone C. It is well known that (FC , #) is a lattice, called the face lattice of C (the partial ordering is setwise containment) see [4], 14] We let, in any lattice, F # G (F # G) denote the smallest upper bound or join (greatest lower bound or meet) of the elements F and G. Corollary 2.5 (P, #) is a lattice which is anti isomorphic to the face lattice (FC , #) Proof. This may be proved directly, but it also follows from ....

A. Br#ndsted. An introduction to convex polytopes. Springer, New York, 1983.

....efficient implementation. 3 Preliminary Review and Results on Polytopes We will now extend the results given in [6] to higher order state spaces. In fact, some of the proofs given herein are not directly applicable to the n = 2 case The notation and results on polytopes largely follows the text [3] by A. Br ndsted. In addition to setting forth the basic notations and results on polytopes, we will develop an important result on polytopes (Theorem 3.27) within this section. We will fix the form of the set U used herein, which are the basis of the representation of the polytopes generated. U ....

A. Brøndsted. An Introduction to Convex Polytopes. Springer-Verlag, New York, 1983.

....This fact does not hold for higher dimensions. Still our results and proofs can be generalized if ext(A) is replaced by cl(ext(A) The most important result on extreme points is by Minkowski (later generalized as the famous Krein Milman theorem of functional analysis) Theorem 2. 8 (Minkowski [2, 7]) Let A be a convex region. Then A is the convex hull of its extreme points: A = conv(ext(A) We will need the following, stronger version of this theorem. For the sake of completeness we include a proof. Theorem 2.9 Let A be a convex region. For every 0 there is a finite subset D of ....

A. Brøndsted. An introduction to convex polytopes. Springer-Verlag, New York, 1983.

....P (m) yields a 2ffi polytope with m ffi vertices and ffim facets, which thus has a face complexity that for fixed dimension is asymptotically worst possible and the same as the one of dual cyclic polytopes. More information on polytopes can be found in the books of Grunbaum [23] Br ndsted [9], and Ziegler [38] 3 Polytope Families In this section we introduce three types of polytopes, fat lattice, intricate, and dwarfed. For each type of polytope we give explicit infinite uniparametric and biparametric families. Recall that a uniparametric family comprises polytopes P of arbitrarily ....

....(i) all surviving vertices of P , and (ii) all points of the form e h, where e is an edge of P . 2. The facet set of P 0 comprises (a) the new facet F 0 = P h, and (b) the old facets F 0 = F H where F ranges over all facets of P that contain some surviving vertex. Proof: See [9], Theorem 11.11. Lemma 7 If P is bounded, then the subgraph of the skeleton of P formed by the surviving vertices and edges is connected. Proof: Define a linear program with the constraints H(P ) f H g and an objective function of the inward normal of H. From the correctness of the simplex ....

A. Brøndsted. Introduction to Convex Polytopes. Springer Verlag, 1981.

....efficient computational schemes. The important advantage is that the geometric connection information is known a priori, and it is ensured that this connection information is preserved. We will now review some basic definitions on polytopes; the notation follows that of the text by A. Br ndsted ([3]) The hyperplane (line) H(u; ff) is the set H(u; ff) fx: hu; xi = ffg. The half plane K(u; ff) is the set K(u; ff) fx: hu; xi ffg. A polyhedral set (in R n ) is the intersection of a finite number of closed half planes in R n , or R n itself. A convex polytope is the convex hull of ....

A. Brøndsted. An Introduction to Convex Polytopes. Springer-Verlag, New York, 1983.

....of S (see [11] Upper Bounds. According to the upper bound theorem for convex polytopes, the maximum numbers of 1 , 2 , and 3 faces of a convex polytope with n 5 vertices in IR 4 are 1 2 (n 2 Gamma n) n 2 Gamma 3n, and 1 2 (n 2 Gamma 3n) respectively (see for example [3, 32]) The lifting map implies the same upper bounds for jF k j, with 1 k 3. As a matter of fact, the upper bound for jF 3 j is one less than for the number of 3 faces of conv(SU ) because at least one 3 face belongs to the upper boundary of conv(SU ) By a result of [38] these bounds are tight ....

A Brønsted. An Introduction to Convex Polytopes. Graduate Texts in Mathematics. SpringerVerlag, New York, 1983.

.... nite set A we let IR A denote the set of real vectors indexed by A (so, by selecting an ordering of the elements of A this set is the Euclidean space IR A ) We use fairly standard graph theoretic notation, see any modern text book in graph theory. For polyhedral theory, we refer to [14] [3] or [12] A#B denotes the symmetric dioeerence of two sets A and B, i.e. A B) # (B A) 2. Approximation and the constrained shortest path problem. We shall describe the curve approximation problem of interest in this paper. Let [a, b] be a nonempty interval of real numbers. For a set of ....

....vertices of M (k) are (i) the vertices of M that satisfy x(E) # k, and (ii) the points obtained as the intersection of the relative interior of an edge of M with the hyperplane x # IR n : x(E) k . This follows from a general result on the intersection of a polytope and a halfspace, see e.g. [3]. Consider an edge F of M having a relative interior point x # in the hyperplane x # IR E : x(E) k . By Proposition 3.2 there is a split path Q = P 1 S P 2 with S = C 1 # C 2 and C 1 C 2 such that F = # Q1 , # Q2 ] where Q i = P 1 C i P 2 for i = 1, 2. Since x # (E) ....

A. Br#ndsted, An introduction to convex polytopes, Springer, New York, 1983.

....[33] allows the definition of linear figures using the constructed solid geometry method. A practical tool to deal with semi linear sets are polytopes. A polytope in the Euclidean space (of arbitrary dimension) is defined as the convex hull of a non empty finite set of points in that space [5, 23, 17]. An open polytope is the topological interior of a polytope with respect to the smallest sub space containing the polytope. It can be proved that bounded semi linear sets and finite unions of open polytopes are equivalent. 33] However, since polytopes are necessarily bounded, finite unions of ....

A. Brøndsted, An Introduction to Convex Polytopes, Graduate Texts in Mathematics, vol. 90, Springer-Verlag, New York, 1983.

....on the n=2 level in E 2 [4] Thus the contribution here is mostly one of observed connections and new proofs, and not new theorems. Section 3 uses ideas of linear programming duality to show that the bound on i minima readily implies the celebrated Upper Bound Theorem for convex polytopes[6, 1]. Here we mean only the upper bound of that theorem, and do not characterize the polytopes for which the bound is tight. 2 The bound for i minima Some preliminary notation: for a set S, let Gamma S k Delta denote the collection of subsets of S of size k, so j Gamma S k Delta j = Gamma ....

....= X i i k g i (P) 2) since each k face F has a unique bottom vertex v, with all k edges in F incident to v pointing up. To bound the quantities f k (P) it is enough to bound g i (P) The above condenses the discussion in Br ndsted s text of McMullen s proof of the Upper Bound Theorem[6, 1]. The LP dual arrangement. The linear programming problem maxfwx j x 2 Pg has the dual problem minfyb j y 2 P 0 g; where P 0 = fy 2 E n j y 2 F ; y 0g; and F = fy 2 E n j yA = wg is an (n Gamma d) flat. Letting d 0 = n Gamma d, the d 0 polytope P 0 is one cell in the ....

A. Brøndsted. An Introduction to Convex Polytopes. Springer-Verlag, Berlin, 1983.

....set, the faces form a lattice ordered by inclusion. Two polytopes are said to be combinatorially equivalent if their face lattices are isomorphic and dual if their face lattices are anti isomorphic (i.e. isomorphic with the direction of inclusion reversed) The following is well known (see e.g. [5]) Proposition 1 If P = conv X is a polytope such that O 2 int P , then Q = f y j Xy 1 g is a polytope dual to P such that O 2 int Q. 2 Primal Dual Algorithms In this section we consider the relationship between the complexity of the primal problem and the complexity of the dual problem for ....

A. Brøndsted. Introduction to Convex Polytopes. Springer-Verlag, 1981.

No context found.

A. Brøndsted, An Introduction to Convex Polytopes, Springer-Verlag, Berlin, New York, 1983.

First 50 documents

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC