Wiki
Wiki

Research notes on every problem, and a library of the papers behind them. Built from the open erdos repository.

Updated


Retain the odd dimension dd and cyclic matrix CmC_m from [[additive_combinatorics/adamczewski_2026_erdos1/corollary_2_2|Corollary 2.2]].

Statement

For a real u0u_0 and integers u1,…,ud−1u_1,\ldots,u_{d-1}, the bound ∥Cmu∥∞<1\|C_mu\|_\infty<1 forces

∣u0+u1+⋯+ud−1∣<1.|u_0+u_1+\cdots+u_{d-1}|<1.

Proof

Abbreviate the integer coordinates as zi=uiz_i=u_i (i≥1i\geq1). For an index jj at which ∣uj∣|u_j| is largest, the inequality ∣2uj+uj−1∣<2|2u_j+u_{j-1}|<2 yields

2∣uj∣≤∣2uj+uj−1∣+∣uj−1∣<2+∣uj∣,2|u_j|\leq|2u_j+u_{j-1}|+|u_{j-1}|<2+|u_j|,

so every coordinate lies in (−2,2)(-2,2) and every ziz_i lies in {−1,0,1}\{-1,0,1\}.

The real coordinate also satisfies ∣u0∣<1|u_0|<1. Suppose u0≥1u_0\geq1; then ∣2u0+zd−1∣<2|2u_0+z_{d-1}|<2 forces zd−1=−1z_{d-1}=-1, while ∣2z1+u0∣<2|2z_1+u_0|<2 forces z1∈{−1,0}z_1\in\{-1,0\}. Define the integer vector

w0=1,wi=zi(1≤i<d).w_0=1,\qquad w_i=z_i\quad(1\leq i<d).

At index 00, ∣2w0+wd−1∣=1|2w_0+w_{d-1}|=1; at index 11, ∣2w1+w0∣=1|2w_1+w_0|=1; and at all remaining indices the original inequalities apply. This contradicts [[additive_combinatorics/adamczewski_2026_erdos1/lemma_2_1|Lemma 2.1]], since w0=1w_0=1. So u0<1u_0<1, and since −u-u satisfies the same hypotheses, also u0>−1u_0>-1.

For each 1≤i<d−11\leq i<d-1, checking the nine possible pairs under

∣2zi+1+zi∣<2,zi,zi+1∈{−1,0,1},|2z_{i+1}+z_i|<2,\qquad z_i,z_{i+1}\in\{-1,0,1\},

shows that zi+1=0z_{i+1}=0 or zi+1=−ziz_{i+1}=-z_i. Once a zero occurs, every later term is zero. Thus the nonzero terms of the tail z1,…,zd−1z_1,\ldots,z_{d-1} form an alternating initial segment z1,…,zpz_1,\ldots,z_p, whose consecutive pairs cancel, so the tail sums to 00 when pp is even and to z1z_1 when pp is odd.

If the tail sums to 00, the total is u0u_0, of absolute value below 11. In the other case the tail sum is z1z_1. If z1=1z_1=1, then ∣2z1+u0∣<2|2z_1+u_0|<2 gives u0<0u_0<0, which together with ∣u0∣<1|u_0|<1 gives −1<u0<0-1<u_0<0 and hence 0<u0+z1<10<u_0+z_1<1. If z1=−1z_1=-1, it gives u0>0u_0>0, hence 0<u0<10<u_0<1 and −1<u0+z1<0-1<u_0+z_1<0. The case z1=0z_1=0 belongs to the first alternative. Therefore the total coordinate sum always has absolute value less than 11.

Source and dependencies

An explanation of the proof of Erdős Problem 1, preliminary exposition with no named author (erdosproblems.com, 2026), §2, Lemma 2.3, pp. 2–3. The edition read is named on the source card. The final two sign cases make explicit the source's last one-line estimate.

Bears on. #1.