FormalConjectures.ErdosProblems.«867»

/-
Copyright 2026 The Formal Conjectures Authors.

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    https://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
-/

import FormalConjecturesUtil

Erdős Problem 867

References:

erdosproblems.com/867

[CoPh96] Coppersmith, Don and Phillips, Steven, On a question of Erdős on subsequence sums. SIAM J. Discrete Math. (1996), 173-177.

[Fr93] Freud, R., Adding numbers - on a problem of P. Erdős. James Cook Mathematical Notes (1993), 6199-6202.

open Filternamespace Erdos867

A finite set of naturals $A={a_1<\cdots<a_t}$ is consecutive-sum-free if it has no solutions to $a_i+a_{i+1}+\cdots+a_j\in A$ with $i<j$; equivalently, whenever an interval $[m,n]$ contains at least two elements of $A$, the sum of the elements of $A$ lying in $[m,n]$ is not itself an element of $A$.

def ConsecutiveSumFree (A : Finset ℕ) : Prop :=
  ∀ m n : ℕ, 2 ≤ (Finset.Icc m n ∩ A).card → (∑ a ∈ Finset.Icc m n ∩ A, a) ∉ A

Is it true that if $A={a_1<\cdots <a_t}\subseteq {1,\ldots,N}$ has no solutions to $$a_i+a_{i+1}+\cdots+a_j\in A$$ then $$\lvert A\rvert \leq \frac{N}{2}+O(1)?$$

In fact this problem is false. Freud [Fr93] constructed a sequence with density $\geq 19/36$. The current best bounds are due to Coppersmith and Phillips [CoPh96], who prove that the maximal size of such an $A$ satisfies $$\frac{13}{24}N -O(1)\leq \lvert A\rvert \leq \left(\frac{2}{3}-\frac{1}{512}\right)N+\log N.$$

@[category research solved, AMS 5 11, formal_proof using lean4 at "https://github.com/plby/lean-proofs/blob/main/src/v4.29.1/ErdosProblems/Erdos867.lean"]
theorem erdos_867 : answer(False) ↔
    ∃ C : ℝ, ∀ N : ℕ, ∀ A ⊆ Finset.Icc 1 N, ConsecutiveSumFree A →
      (A.card : ℝ) ≤ (N : ℝ) / 2 + C := by⊢ False ↔ ∃ C, ∀ (N : ℕ), ∀ A ⊆ Finset.Icc 1 N, ConsecutiveSumFree A → ↑A.card ≤ ↑N / 2 + C
  sorryAll goals completed! 🐙

Taking $A=(N/2,N]\cap \mathbb{N}$ shows $\lvert A\rvert \geq N/2-O(1)$ is possible.

@[category research solved, AMS 5 11]
theorem erdos_867.variants.lower_bound :
    ∃ C : ℝ, ∀ N : ℕ, ∃ A ⊆ Finset.Icc 1 N, ConsecutiveSumFree A ∧
      ((N : ℝ) / 2 - C ≤ (A.card : ℝ)) := by⊢ ∃ C, ∀ (N : ℕ), ∃ A ⊆ Finset.Icc 1 N, ConsecutiveSumFree A ∧ ↑N / 2 - C ≤ ↑A.card
  sorryAll goals completed! 🐙

Adenwalla has observed that $$\lvert A\rvert \leq (\tfrac{2}{3}+o(1))N.$$

@[category research solved, AMS 5 11]
theorem erdos_867.variants.adenwalla (ε : ℝ) (hε : 0 < ε) :
    ∀ᶠ N : ℕ in atTop, ∀ A ⊆ Finset.Icc 1 N, ConsecutiveSumFree A →
      (A.card : ℝ) ≤ (2 / 3 + ε) * (N : ℝ) := byε:ℝhε:0 < ε⊢ ∀ᶠ (N : ℕ) in atTop, ∀ A ⊆ Finset.Icc 1 N, ConsecutiveSumFree A → ↑A.card ≤ (2 / 3 + ε) * ↑N
  sorryAll goals completed! 🐙

Freud [Fr93] constructed a sequence with density $\geq 19/36$.

@[category research solved, AMS 5 11]
theorem erdos_867.variants.freud :
    ∀ ε : ℝ, 0 < ε → ∀ᶠ N : ℕ in atTop, ∃ A ⊆ Finset.Icc 1 N, ConsecutiveSumFree A ∧
      (((19 / 36 : ℝ) - ε) * (N : ℝ) ≤ (A.card : ℝ)) := by⊢ ∀ (ε : ℝ), 0 < ε → ∀ᶠ (N : ℕ) in atTop, ∃ A ⊆ Finset.Icc 1 N, ConsecutiveSumFree A ∧ (19 / 36 - ε) * ↑N ≤ ↑A.card
  sorryAll goals completed! 🐙

The current best bounds are due to Coppersmith and Phillips [CoPh96], who prove that the maximal size of such an $A$ satisfies $$\frac{13}{24}N -O(1)\leq \lvert A\rvert.$$

@[category research solved, AMS 5 11]
theorem erdos_867.variants.coppersmith_phillips_lower_bound :
    ∃ C : ℝ, ∀ N : ℕ, ∃ A ⊆ Finset.Icc 1 N, ConsecutiveSumFree A ∧
      ((13 / 24 : ℝ) * (N : ℝ) - C ≤ (A.card : ℝ)) := by⊢ ∃ C, ∀ (N : ℕ), ∃ A ⊆ Finset.Icc 1 N, ConsecutiveSumFree A ∧ 13 / 24 * ↑N - C ≤ ↑A.card
  sorryAll goals completed! 🐙

The current best bounds are due to Coppersmith and Phillips [CoPh96], who prove that the maximal size of such an $A$ satisfies $$\lvert A\rvert \leq \left(\frac{2}{3}-\frac{1}{512}\right)N+\log N.$$

@[category research solved, AMS 5 11]
theorem erdos_867.variants.coppersmith_phillips_upper_bound :
    ∀ᶠ N : ℕ in atTop, ∀ A ⊆ Finset.Icc 1 N, ConsecutiveSumFree A →
      (A.card : ℝ) ≤ (2 / 3 - 1 / 512) * (N : ℝ) + Real.log (N : ℝ) := by⊢ ∀ᶠ (N : ℕ) in atTop, ∀ A ⊆ Finset.Icc 1 N, ConsecutiveSumFree A → ↑A.card ≤ (2 / 3 - 1 / 512) * ↑N + Real.log ↑N
  sorryAll goals completed! 🐙

end Erdos867