On the complexity of the independent set problem in triangle graphs

We consider the complexity of the maximum (maximum weight) independent set problem within triangle graphs, i.e., graphs G satisfying the following triangle condition: for every maximal independent set I in G and every edge uv in G−I, there is a vertex w∈I such that {u,v,w} is a triangle in G. We also introduce a new graph parameter (the upper independent neighborhood number) and the corresponding upper independent neighborhood set problem. We show that for triangle graphs the new parameter is equal to the independence number. We prove that the problems under consideration are NP-complete, even for some restricted subclasses of triangle graphs, and provide several polynomially solvable cases for these problems within triangle graphs. Furthermore, we show that, for general triangle graphs, the maximum independent set problem and the upper independent neighborhood set problem cannot be polynomially approximated within any fixed constant factor greater than one unless P=NP.     

Analysis of parallel algorithms for finding a maximal independent set in a random hypergraph

It is well known [9] that finding a maximal independent set in a graph is in class NC and [10] that finding a maximal independent set in a hypergraph with fixed dimension is in RNC. It is not known whether this latter problem remains in NC when the dimension is part of the input. We will study the problem when the problem instances are randomly chosen. It was shown in [6] that the expected running time of a simple parallel algorithm for finding the lexicographically first maximal independent set (Ifmis) in a random simple graph is logarithmic in the input size. In this paper, we will prove a generalization of this result. We show that if a random k-uniform hypergraph has vertex set {1, 2, …, n} and its edges are chosen independently with probability p from the set of (ⁿ_k) possible edges, then our algorithm finds the Ifmis in O( ) expected time. The hidden constant is independent of k, p. © 1996 John Wiley & Sons, Inc. Random Struct. Alg., 9 , 359–377 (1996)     

Dichotomies of the set of test measures of a Haar-null set

We prove that ifX is a Polish space andF a face ofP(X) with the Baire property, thenF is either a meager or a co-meager subset ofP(X). As a consequence we show that for every abelian Polish groupX and every analytic Haar-null set Λ⊆X, the set of test measuresT(Λ) of Λ is either meager or co-meager. We characterize the non-locally-compact groups as the ones for which there exists a closed Haar-null setF⊆X withT(F) meager, Moreover, we answer negatively a question of J. Mycielski by showing that for every non-locally-compact abelian Polish group and every σ-compact subgroupG ofX there exists aG-invariantF _σ subset ofX which is neither prevalent nor Haar-null. Research supported by a grant of EPEAEK program “Pythagoras”.     

Finding a minimum independent dominating set in a permutation graph

We give a linear time reduction of the problem of finding a minimum independent dominating set in a permutation graph, into that of finding a shortest maximal increasing subsequence. We then give an O(n log²n)-time algorithm for solving the second (and hence the first) problem. This improves on the O(n³)-time algorithm given in [4] for solving the problem of finding a minimum independent dominating set in a permutation graph.     

A note on cliques and independent sets

For integers m, n ≥ 2, let f (m, n) be the minimum order of a graph where every vertex belongs to both a clique of cardinality m and an independent set of cardinality n. We show that . © 1997 John Wiley & Sons, Inc.     

Maximal independent sets in bipartite graphs

A maximal independent set of a graph G is an independent set that is not contained properly in any other independent set of G. In this paper, we determine the maximum number of maximal independent sets among all bipartite graphs of order n and the extremal graphs as well as the corresponding results for connected bipartite graphs. © 1993 John Wiley & Sons, Inc.     

Maximal matchings in graphs with large neighborhoods of independent vertices

We obtain lower bounds on the size of a maximum matching in a graph satisfying the condition |N(X)| ≥ s for every independent set X of m vertices, thus generalizing results of Faudree, Gould, Jacobson, and Schelp for the case m = 2.     

Fast local search for the maximum independent set problem

Given a graph G=(V,E), the independent set problem is that of finding a maximum-cardinality subset S of V such that no two vertices in S are adjacent. We introduce two fast local search routines for this problem. The first can determine in linear time whether a maximal solution can be improved by replacing a single vertex with two others. The second routine can determine in O(mΔ) time (where Δ is the highest degree in the graph) whether there are two solution vertices than can be replaced by a set of three. We also present a more elaborate heuristic that successfully applies local search to find near-optimum solutions to a wide variety of instances. We test our algorithms on instances from the literature as well as on new ones proposed in this paper.     

Another note on cliques and independent sets

Entringer, Goddard, and Henning studied graphs in which every vertex belongs to both an (m + 1)‐clique and an independent (n + 1)‐set; they proved that there is such a graph of order p if and only if . We give an alternative and slightly easier proof of this fact, relating it to combinatorial matrix theory. © 2000 John Wiley & Sons, Inc. J Graph Theory 35: 173–175, 2000     

Almost optimal set covers in finite VC-dimension

We give a deterministic polynomial-time method for finding a set cover in a set system (X, ℛ) of dual VC-dimensiond such that the size of our cover is at most a factor ofO(d log(dc)) from the optimal size,c. For constant VC-dimensional set systems, which are common in computational geometry, our method gives anO(logc) approximation factor. This improves the previous Θ(log⋎X⋎) bound of the greedy method and challenges recent complexity-theoretic lower bounds for set covers (which do not make any assumptions about the VC-dimension). We give several applications of our method to computational geometry, and we show that in some cases, such as those arising in three-dimensional polytope approximation and two-dimensional disk covering, we can quickly findO(c)-sized covers. The first author was supported in part by NSF Grant CCR-90-02352 and Ecole Normale Supérieure. The second author's research was supported by the NSF and DARPA under Grant CCR-8908092, and by the NSF under Grants IRI-9116843 and CCR-9300079.     

Maximal sets of triangle-factors on v = 6m vertices

We show that the necessary condition m ≤ k ≤ 3m − 1 that there exists a maximal set of k triangle-factors on 6m ≥ 18 vertices is also sufficient, except possibly when k = m. © 1998 John Wiley & Sons, Inc. J Combin Designs 6: 235–244, 1998     

Facets of the independent set polytope

Theoretical results pertaining to the independent set polytope P^ISP=conv{x{0,1}ⁿ:Axb} are presented. A conflict hypergraph is constructed based on the set of dependent sets which facilitates the examination of the facial structure of P^ISP. Necessary and sufficient conditions are provided for every nontrivial 0-1 facet-defining inequalities of P^ISP in terms of hypercliques. The relationship of hypercliques and some classes of knapsack facet-defining inequalities are briefly discussed. The notion of lifting is extended to the conflict hypergraph setting to obtain strong valid inequalities, and back-lifting is introduced to strengthen cut coefficients. Preliminary computational results are presented to illustrate the usefulness of the theoretical findings.Mathematics Subject Classification (2000): 90C11, 90C57, 90C35     

A class of planar well-covered graphs with girth four

A well-covered graph is a graph in which every maximal independent set is a maximum independent set; Plummer introduced the concept in a 1970 paper. The notion of a 1-well-covered graph was introduced by Staples in her 1975 dissertation: a well-covered graph G is 1-well-covered if and only if G - v is also well covered for every point v in G. Except for K₂ and C₅, every 1-well-covered graph contains triangles or 4-cycles. We show that all planar 1-well-covered graphs of girth 4 belong to a specific infinite family, and we give a characterization of this family. © 1995 John Wiley & Sons, Inc.     

Fragile graphs with small independent cuts

A graph is called fragile if it has a vertex cut which is also an independent set. Chen and Yu proved that every graph with n vertices and at most 2n?4 edges is fragile, which was conjectured to be true by Caro. However, their proof does not give any information on the number of vertices in the independent cuts. The purpose of this paper is to investigate when a graph has a small independent cut. We show that if G is a graph on n vertices and at most (12n/7)?3 edges, then G contains an independent cut S with ∣S∣≤3. Upper bounds on the number of edges of a graph having an independent cut of size 1 or 2 are also obtained. We also show that for any positive integer k, there is a positive number ε such that there are infinitely many graphs G with n vertices and at most (2?ε)n edges, but G has no independent cut with less than k vertices. © 2002 Wiley Periodicals, Inc. J Graph Theory 41: 327–341, 2002     

A note on the random greedy independent set algorithm

Let r be a fixed constant and let be an r‐uniform, D‐regular hypergraph on N vertices. Assume further that for some . Consider the random greedy algorithm for forming an independent set in . An independent set is chosen at random by iteratively choosing vertices at random to be in the independent set. At each step we chose a vertex uniformly at random from the collection of vertices that could be added to the independent set (i.e. the collection of vertices v with the property that v is not in the current independent set I and contains no edge in ). Note that this process terminates at a maximal subset of vertices with the property that this set contains no edge of ; that is, the process terminates at a maximal independent set. We prove that if satisfies certain degree and codegree conditions then there are vertices in the independent set produced by the random greedy algorithm with high probability. This result generalizes a lower bound on the number of steps in the H‐free process due to Bohman and Keevash and produces objects of interest in additive combinatorics. © 2016 Wiley Periodicals, Inc. Random Struct. Alg., 49, 479–502, 2016     

Trees with the second largest number of maximal independent sets

A maximal independent set is an independent set that is not a proper subset of any other independent set. In this paper, we determine the second largest number of maximal independent sets among all trees and forests of order n≥4. We also characterize those extremal graphs achieving these values.     

Computability in structures representing a Scott set

Continuing work begun in [10], we utilize a notion of forcing for which the generic objects are structures and which allows us to determine whether these “generic” structures compute certain sets and enumerations. The forcing conditions are bounded complexity types which are consistent with a given theory and are elements of a given Scott set. These generic structures will “represent” this given Scott set, in the sense that the structure has a certain weak saturation property with respect to bounded complexity types in the Scott set. For example, if ? is a nonstandard model of PA, then ? represents the Scott set ? = n∈ω | ?⊧“the nth prime divides a” | a∈?. The notion of forcing yields two main results. The first characterizes the sets of natural numbers computable in all models of a given theory representing a given Scott set. We show that the characteristic function of such a set must be enumeration reducible to a complete existential type which is consistent with the given theory and is an element of the given Scott set. The second provides a sufficient condition for the existence of a structure ? such that ? represents a countable jump ideal and ? does not compute an enumeration of a given family of sets ?. This second result is of particular interest when the family of sets which cannot be enumerated is ? = Rep[Th(?)]. Under this additional assumption, the second result generalizes a result on TA [6] and on certain other completions of PA [10]. For example, we show that there also exist models of completions of ZF from which one cannot enumerate the family of sets represented by the theory. Received: 8 October 1997 / Published online: 25 January 2001     

Maximum weight independent set in trees

Computing a maximum independent set, weighted or unweighted, isNP-hard for general as well as planar graphs. However, polynomial time algorithms do exist for solving this problem on special classes of graphs. In this paper we present an efficient algorithm for computing a maximum weight independent set in trees. A divide and conquer approach based on centroid decomposition of trees is used to compute a maximum weight independent set withinO(n logn) time, wheren is the number of vertices in the tree. We introduce a notion of analternating tree which is crucial in obtaining a new independent set from the previous one.     

The construction of a set of recurrence which is not a set of strong recurrence

This paper deals with two possible definitions of recurrence in measure preserving systems. A set of integersR is said to be a set of (Poincaré) recurrence if, for all measure preserving systems (X, B, μ, T) and any measurable setA of positive measure, there is anr εR such thatμ(T ^r A∩A)>0.R is said to be a set of strong recurrence if, for all measure preserving systems (X, B, μ, T) and any measurable setA of positive measure, there is ane>0 and an infinite number of elementsr ofR such thatμ(T ^r A∩A)≥e (see Bergelson’s 1985 paper). This paper constructs a set of recurrenceR, an example of a measure preserving system (X, B, μ, T) and a measurable setA of measure 1/2, such that lim _{r→∞:rεRμ} (A∩T ^r A)=0. In particularR is a set of recurrence but not a set of strong recurrence, giving a negative answer to a question of Bergelson posed in 1985. Further, it also constructs a set of recurrence which does not force the continuity of positive measures and so reproves a result of Bourgain published in 1987. This paper forms a part of the author’s Ph.D. Thesis at the Ohio State University. The author wishes to thank his advisor, Professor Bergelson, for suggesting the problem of this paper and for his guidance.     

No large sets which can be translated away from every Marczewski null set

There is no set of size continuum which is “s₀-shiftable”, i. e., which can be translated away from every set in the Marczewski ideal s₀ (where a set of reals is in s₀ if for every perfect set there is a perfect subset disjoint from it). (© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)