We study the problems of adjacency sketching, small-distance sketching, and approximate distance threshold sketching for monotone classes of graphs. The problem is to obtain randomized sketches of the vertices of any graph G in the class, so that adjacency, exact distance thresholds, or approximate distance thresholds of two vertices u, v can be decided (with high probability) from the sketches of u and v, by a decoder that does not know the graph. The goal is to determine when sketches of constant size exist. We show that, for monotone classes of graphs, there is a strict hierarchy: approximate distance threshold sketches imply small-distance sketches, which imply adjacency sketches, whereas the reverse implications are each false. The existence of an adjacency sketch is equivalent to the condition of bounded arboricity, while the existence of small-distance sketches is equivalent to the condition of bounded expansion. Classes of constant expansion admit approximate distance threshold sketches, while a monotone graph class can have arbitrarily small non-constant expansion without admitting an approximate distance threshold sketch.
翻译:我们研究相近草图、小距离草图和单调类图的近距离起点草图的问题。 问题在于获取该类中任何图形G的脊椎的随机草图,以使两个脊椎的相近性、准确的距离阈值或近似距离阈值u, v 可以通过一个不知道图形的分解器从u 和 v 的草图中决定(高概率), 目标是确定何时存在常数大小的草图。 我们显示,对于单调类图来说,有严格的等级: 近似距离门槛草图意味着小距离草图,这意味着相近性草图,而反面的影响则都是假的。 相近性草图的存在相当于受约束的偏差性,而小距离草图的存在则相当于受约束的扩展条件。 持续扩展的类别承认近似距离阈值的素描,而单调的单调式图类则可以任意地进行小距离非科的扩展,而无需设定近似距离的草图。