How statement and proof provenance work
The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.
- Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
- AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
- AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.
These labels describe origin, not correctness: citations and verification chips remain separate evidence.
Equitable strong regularity lemma: a very regular refinement that changes energy only slightly
Statement
Let and let . There is such that every finite graph with has equitable partitions satisfying
- refines ;
- is -regular;
- is -regular;
- ; and
- and .
Facts & Assumptions
Given: A nonincreasing positive sequence , an integer , and a finite graph of order .
For every and there is such that every partition into at most nonempty parts has an -regular refinement into at most parts, which may be chosen equitable when the given partition is equitable (A prescribed finite vertex partition has a bounded -regular refinement, equitable when the initial partition is equitable).
For every and there is such that every graph of order at least has an equitable -regular partition into parts with (Szemerédi regularity lemma with an equitable partition and an explicit tower-type upper bound for graphs of order at least ).
Energy is nondecreasing under refinement and lies in (Energy lies in and cannot decrease under refinement).
An equitable partition has part sizes differing by at most one, and a partition is -regular when its irregular ordered pairs carry total weight at most (-regular vertex partitions, equitable partitions, and refinement).
is -regular when every and with and satisfies (-regular pairs and self-regular vertex sets).
Proof
If then every -regular pair is -regular, because the subsets tested at parameter are among those tested at ; by [L4] the same monotonicity passes to partitions. So replacing each by keeps the sequence positive and nonincreasing and only strengthens every conclusion, and we may assume for all . Let be the constant of [L2].
Small graphs. Suppose and let be the partition of into singletons. It refines itself, is equitable, and has parts. The only subset of a singleton of size at least times its size is the singleton itself, so by [L5] every pair of singletons is -regular for every , giving conclusions 2 and 3; and gives conclusion 4.
Large graphs. Suppose . Use [L2] to choose an equitable -regular partition with . Having constructed , apply [L1] to with parameter to choose an equitable refinement that is -regular.
Each part count is bounded by a function of the preceding one and the fixed sequence, so for every fixed number of stages all are bounded independently of . A refinement has at least as many parts as the partition it refines, so for every .
Set . If for every , telescoping would give , contrary to [L3]. Hence some satisfies ; put and .
Step 2.2 supplies refinement, equitability, and -regularity of . Also is -regular: this holds for by construction, and every later is -regular with , which gives -regularity by step 1.1. Step 3.1 gives and step 3.2 the energy bound.
Let be the largest of and the recursively obtained part-count bounds through stage . Step 2.1 settles and steps 3.1 and 4.1 settle , so in both cases and all five conclusions hold.
Depends on
- A prescribed finite vertex partition has a bounded $\epsilon$-regular refinement, equitable when the initial partition is equitable
- Szemerédi regularity lemma with an equitable partition and an explicit tower-type upper bound for graphs of order at least $m_0$
- Energy lies in $[0,1]$ and cannot decrease under refinement
- $\epsilon$-regular vertex partitions, equitable partitions, and refinement
- $\epsilon$-regular pairs and self-regular vertex sets
Used by
Dependency tree · next 3 levels
Direct dependencies and their dependencies through the next three levels: 15 results over 9 levels. An arrow runs from a result to what uses it, and this result sits at the bottom with a heavier outline. Click the chart to enlarge it.
Sources
- Y. Zhao, Graph Theory and Additive Combinatorics, Theorem 2.8.3 and Remark 2.8.6 (standard reference, not scraped)
- D. Conlon and J. Fox, Graph removal lemmas, sec. 2.3 (standard reference, not scraped)