arXiv subjects

Daniel C. McDonald

Publications and source records attributed to Daniel C. McDonald.

5 recordsLinked to original sources

arXiv2026

Optimally reconnecting graphs against an edge-destroying adversary

We introduce a model involving two adversaries Buster and Fixer taking turns modifying a connected graph, where each round consists of Buster deleting a subset of edges and Fixer responding by adding edges from a finite reserve set of weighted edges to leave the graph connected, with Buster limited by the total number of edges he is allowed to delete throughout the game. Fixer wins if she can reconnect the graph after Buster has reached his limit of edges to delete, while Buster wins if he can delete edges in such a way that Fixer cannot reconnect the graph using the remaining edges in reserve. With the weights representing the cost for Fixer to use specific reserve edges to reconnect the graph, we prove that a greedy strategy for Fixer always results in an optimal result for Fixer: victory, if possible, for as cheaply as can be guaranteed against any Buster strategy, and if defeat cannot be avoided, the cheapest possible loss that can be guaranteed against any Buster strategy.

math.CO↗

arXiv2015

Connectedness and Hamiltonicity of graphs on vertex colorings

Given a graph $H$, let $G^j_k(H)$ be the graph whose vertices are the proper $k$-colorings of $H$, with edges joining two colorings if $H$ contains a connected subgraph on at most $j$ vertices that includes all vertices where the colorings differ. Properties of $G^1_k(H)$ have been investigated before, including connectedness and Hamiltonicity. We introduce and study the parameters $g_k(H)$ and $h_k(H)$, which denote the minimum $j$ such that $G^j_k(H)$ is connected or Hamiltonian, respectively.

math.CO↗

arXiv2014

List rankings and on-line list rankings of graphs

A $k$-ranking of a graph $G$ is a labeling of its vertices from $\{1,\ldots,k\}$ such that any nontrivial path whose endpoints have the same label contains a larger label. The least $k$ for which $G$ has a $k$-ranking is the ranking number of $G$, also known as tree-depth. The list ranking number of $G$ is the least $k$ such that if each vertex of $G$ is assigned a set of $k$ potential labels, then $G$ can be ranked by labeling each vertex with a label from its assigned list. Rankings model a certain parallel processing problem in manufacturing, while the list ranking version adds scheduling constraints. We compute the list ranking number of paths, cycles, and trees with many more leaves than internal vertices. Some of these results follow from stronger theorems we prove about on-line versions of list ranking, where each vertex starts with an empty list having some fixed capacity, and potential labels are presented one by one, at which time they are added to the lists of certain vertices; the decision of which of these vertices are actually to be ranked with that label must be made immediately.

math.CO↗

arXiv2014

On-line vertex ranking of trees

A $k$-ranking of a graph $G$ is a labeling of its vertices from $\{1,\ldots,k\}$ such that any nontrivial path whose endpoints have the same label contains a larger label. The least $k$ for which $G$ has a $k$-ranking is the ranking number of $G$, also known as tree-depth. Applications of rankings include VLSI design, parallel computing, and factory scheduling. The on-line ranking problem asks for an algorithm to rank the vertices of $G$ as they are presented one at a time along with all previously ranked vertices and the edges between them (so each vertex is presented as the lone unranked vertex in a partially labeled induced subgraph of $G$ whose final placement in $G$ is not specified). The on-line ranking number of $G$ is the minimum over all such algorithms of the largest label that algorithm can be forced to use. We give bounds on the on-line ranking number of trees in terms of maximum degree, diameter, and number of interior vertices.

math.CO↗

arXiv2014

A combinatorial proof on partition function parity

One of the most basic results concerning the number-theoretic properties of the partition function $p(n)$ is that $p(n)$ takes each value of parity infinitely often. This statement was first proved by Kolberg in 1959, and it was strengthened by Subbarao in 1966 to say that both $p(2n)$ and $p(2n+1)$ take each value of parity infinitely often. These results have received several other proofs, each relying to some extent on manipulating generating functions. We give a new, self-contained proof of Subbarao's result by constructing a series of bijections and involutions, along the way getting a more general theorem concerning the enumeration of a special subset of integer partitions.

math.NT↗