sink (graph theory)
This article was accepted into the corpus but its outbound wikilinks were never NER-processed — typical at the deepest BFS hop or when the run's entity cap was reached. No expansion funnel to show.
| sink (graph theory) | |
|---|---|
.png) | |
| Name | sink (graph theory) |
| Caption | A directed graph with a sink vertex highlighted |
| Type | Graph theory concept |
| Field | Mathematics |
| Related | source (graph theory), strongly connected component, tournament (graph theory) |
sink (graph theory)
A sink in graph theory is a vertex in a directed graph with no outgoing edges; it is a terminal point for all incident arcs. In finite directed graphs such as tournaments, DAGs, and flow networks the notion of a sink interacts with concepts from combinatorics, algebraic graph theory, and algorithmic complexity. Sinks appear in classical results and structures studied by researchers connected to institutions like Princeton University, Massachusetts Institute of Technology, University of Cambridge, Stanford University, and University of California, Berkeley.
Definition and basic properties
A sink is formally a vertex v in a directed graph G = (V, E) for which every edge incident to v is incoming, equivalently the outdegree of v is zero. Basic properties link sinks to notions from spectral graph theory, order theory, and category theory studied at Harvard University, California Institute of Technology, University of Oxford, Imperial College London, and ETH Zurich. In finite graphs, sinks are distinct from sources, and a vertex may be a sink in one orientation and not in another, a fact relevant to research groups at University of Chicago, Columbia University, and New York University. The relation between sinks and sinks inside strongly connected components connects to work by researchers affiliated with Carnegie Mellon University, University of Illinois Urbana-Champaign, and University of Toronto.
Examples and special cases
Common examples include sinks in directed acyclic graphs (DAGs), tournaments, and bipartite orientations studied in seminars at University of Michigan, University of Washington, Cornell University, and Duke University. In a DAG arising from precedence constraints in scheduling problems from IBM, Microsoft Research, and Google Research, sinks correspond to tasks with no successors; in tournament graphs considered by scholars at Princeton University, Yale University, and Brown University a sink may or may not exist depending on the orientation. In flow networks inspired by works at Bell Labs, AT&T, and Siemens, sinks model sinks of flow distinct from super-sinks introduced by engineers at General Electric. Special cases include universal sink (vertex adjacent from all other vertices) and isolated sink in sparse graphs, topics explored in collaborations involving Stanford University and EPFL.
Existence and characterization in directed graphs
Existence theorems for sinks depend on structural constraints; every finite nonempty DAG has at least one sink, a statement taught in courses at University of Cambridge and Oxford University. Characterizations relate sinks to minimal elements in partial orders studied in seminars at Princeton University and Harvard University and to kernel concepts analyzed by researchers from Tel Aviv University and Technion – Israel Institute of Technology. In tournaments, Landau-type inequalities and score sequences investigated by mathematicians at University of Göttingen and University of Bonn inform when a sink can exist. In random graph models developed at Bell Labs and Los Alamos National Laboratory, expected counts of sinks have been computed by teams at University of California, Los Angeles and University of Maryland.
Algorithms for detection and computation
Algorithms to find sinks range from linear-time scans to reductions to reachability and strongly connected component decomposition, methods refined by algorithm groups at Massachusetts Institute of Technology, École Polytechnique, University of Toronto, and Carnegie Mellon University. For adjacency-matrix representations, a classic linear-time algorithm identifies a universal sink using pointer elimination, an approach taught at Princeton University and McGill University. For large-scale networks, parallel and distributed algorithms from researchers at Microsoft Research, Google Research, Facebook AI Research, and Amazon address sink detection across clusters. Complexity analyses drawing on results from Stanford University and University of California, Berkeley compare worst-case bounds with average-case behavior in Erdős–Rényi and preferential attachment models studied at Institute for Advanced Study and Santa Fe Institute.
Applications and relevance in graph theory
Sinks are relevant in proofs and constructions across combinatorics, fixed-point theorems, and game theory studied at Princeton University and Yale University. They serve as termination witnesses in reduction systems investigated at Cornell University and Brown University and as absorption states in Markov chains analyzed by researchers at Columbia University and Johns Hopkins University. In social choice and voting theory explored at Harvard University and University of Oxford, sinks model candidates who receive no outgoing preferences. In algorithmic applications involving topological sorting, constraint satisfaction, and model checking, work from Microsoft Research, IBM Research, and Google Research employs sinks for optimization and correctness arguments.
Variants and related concepts
Variants include universal sinks, maximal sinks in partial orders, and sinks within strongly connected components; these relate to kernels, sinks in hypergraphs, and sink-finding in dynamic graphs studied by groups at ETH Zurich, EPFL, University of Edinburgh, and Imperial College London. Related concepts include sources, sinks in flow theory, absorbing states in stochastic processes, and minimal elements in posets—topics investigated in collaborations involving Princeton University, University of California, Berkeley, and University of Michigan. Further connections reach into category-theoretic limits, end vertices in undirected graphs studied at University of Warwick, and equilibrium concepts in economic theory analyzed at London School of Economics.