Alphabeta Math
CorollaryStatement: AI-adaptedProof: AI-generated
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.

Destroying cycle spurious minima by bounded-degree edge additions

PDF p. 9: Section 1.5, paragraph on adding edges to a long cycle

Statement

Assume countable choice. For every sufficiently large integer n, any cycle Cn on n vertices can be augmented by adding exactly 598 new edges incident to each vertex so that the resulting graph is globally synchronizing and has no spurious local minima.

Here Cn is a simple undirected graph whose vertices have a cyclic ordering, with edges precisely between consecutive vertices, including the last and first. A graph is 600-regular if every vertex has degree 600. A Hamiltonian cycle H is a cycle containing every vertex. Global synchrony means that, for phase-volume almost every initial state, the rotating-frame homogeneous Kuramoto trajectory converges to a state with all phases equal. A spurious local minimum is a local minimum of EG(θ)=12∑x,yAxy(1−cos⁡(θx−θy)) on the phase torus (R/2πZ)V whose phases are not all equal; A is the graph's adjacency matrix.

More precisely, for a uniformly random labelled simple 600-regular graph on n vertices, the event that it is both Hamiltonian and globally synchronizing has probability tending to one as n→∞ through feasible orders. On that event, its edge set is S⊔H, where each vertex is incident to exactly 598 edges of S. Relabelling H as the prescribed Cn gives the augmentation. Thus the article's bound of at most 600 added edges per vertex holds, in fact with 598.

Facts & Assumptions

[F1]

Under countable choice, a uniformly random labelled simple 600-regular graph is globally synchronizing with probability tending to one as its feasible order tends to infinity (Ramanujan and typical regular graphs synchronize for degree at least 600, Uniform random regular graph convention, The Axiom of Countable Choice (AC_ω)).

[F2]

A uniformly random labelled simple 600-regular graph is Hamiltonian with probability tending to one along the same feasible orders (Typical regular graphs are Hamiltonian, Uniform random regular graph convention).

[F3]
[F4]

Under countable choice, a cycle of length greater than five is not globally synchronizing (Trees synchronize whereas cycles longer than five do not).

[F5]

For any two events B1,B2 in one probability space, P(B1∪B2)≤P(B1)+P(B2) (The finite union bound).

[F6]

Under countable choice, the homogeneous Kuramoto equation in local lifts is θ˙x=ω−∑yAxysin⁡(θx−θy) and has a unique trajectory through each initial state for all real time. Its rotating-frame trajectory is obtained by subtracting ωt from each coordinate (Graph Kuramoto differential equations, Globally synchronizing graph — Definition 1.1, The Axiom of Countable Choice (AC_ω)).

[F7]

Under countable choice, normalized phase volume on TV is computed by (2π)−∣V∣ times Lebesgue measure of representatives in (−π,π]V, independently of the ordering of V (Normalized phase volume and almost-everywhere assertions, The Axiom of Countable Choice (AC_ω)).

Proof

Given: Countable choice and a prescribed cycle Cn of sufficiently large order.

Goal: Find the stated edge augmentation and prove its synchronization and absence of spurious local minima.

1.1givenF1F2F5algebra

Let B1 and B2 be the failure events for synchronization and Hamiltonicity, respectively, in the same uniform random regular graph space. By [F1], [F2] and [F5], P(B1∪B2)≤P(B1)+P(B2)→0. Hence the simultaneous event has probability tending to one and, for all sufficiently large feasible n, has positive probability. Every n≥601 is feasible: on vertices modulo n, joining vertices whose cyclic distance is one of 1,…,300 gives a simple 600-regular graph. Thus, for every sufficiently large integer n, the simultaneous event is nonempty, and at least one simple 600-regular graph G has both properties.

2.1F1F2step 1.1construct

Choose a Hamiltonian cycle H in G and let S=E(G)∖E(H). Each vertex has two incident edges in H and 600 in G, so it has exactly 598 incident edges in S. A bijection σ:V→W preserving cyclic order maps H to the prescribed Cn on W. Apply it to every edge of G to obtain G′; the image of S consists of new edges disjoint from Cn, with exactly 598 incident to each vertex. The adjacency matrices obey Aσ(x),σ(y)′=Axy.

3.1F6step 2.1constructalgebra

Define P:TV→TW by (Pθ)σ(x)=θx. It is well defined modulo 2π, with inverse induced by σ−1. If θ(t) is a homogeneous Kuramoto trajectory for G, then ψ(t)=Pθ(t) satisfies ψ˙σ(x)=ω−∑y∈VAσ(x),σ(y)′sin⁡(ψσ(x)−ψσ(y)). Reindexing this finite sum over W gives exactly the equation for G′. Its initial state is Pθ(0), so uniqueness in [F6] identifies it with that graph's trajectory. Since P commutes with subtracting ωt from every coordinate, it also conjugates the rotating-frame trajectories.

4.1F1F6F7step 1.1step 3.1

Order V arbitrarily and order W by transporting that ordering through σ. In these coordinate identifications P leaves the representative tuple in (−π,π]n unchanged. By [F7], it therefore carries phase-measurable sets to phase-measurable sets with the same normalized volume. The same coordinate identification shows that P is an isometry for the flat torus distance and carries fully synchronized states to fully synchronized states. Let N be the phase-null exceptional set for G. For each initial state outside P(N), its inverse image lies outside N, so its rotating-frame trajectory for G converges to a fully synchronized point. By step 3.1 and the isometry, the corresponding trajectory for G′ converges to the image of that point. Since P(N) is phase-null, G′ is globally synchronizing.

5.1F3F4step 2.1step 4.1∎

If the augmented graph had a spurious local minimum, [F3] would rule out its global synchrony, contradicting step 4.1. Thus it has no spurious local minima. For n>5, [F4] also shows that the original cycle was not globally synchronizing. This proves all assertions.

Source notes

The article's Section 1.5 states the long-cycle edge-addition claim and describes the partition into S and a Hamiltonian cycle H. The source reference records attribution; the proof uses the declared suppliers. The random assertion concerns the simultaneous event before relabelling, and the existence assertion applies to every sufficiently large order.

Depends on

Dependency tree · 2 levels

Each result is drawn at its shortest distance from this item. Arrows run from prerequisites to the items that use them; this item has a heavier outline. External source nodes link directly to the cited text. The view shows up to two dependency steps; open a prerequisite to continue.

Sources