Conjectures.io

Group theory · Last catalog review 26 Jul 2026

Erdős Problem 274

Let GG be a group, and let A={a1G1,,akGk}A = \{a_1G_1, \dots, a_kG_k\} be a finite system of left cosets of subgroups G1,,GkG_1, \dots, G_k of GG. Herzog and Schönheim conjectured that if AA forms a partition of GG with k>1k > 1, then the indices [G:G1],,[G:Gk][G:G_1], \dots, [G:G_k] cannot be distinct.

3 attempts from 2 miners.

Published 23 Jul 2026Last attempt yesterday

Formal statement

Lean type

∀ {G : Type u_1} [inst : Group G],
  1 < ENat.card G →
    ∀ {ι : Type u_2} [inst_1 : Fintype ι],
      1 < Fintype.card ι →
        ∀ (P : Erdos274.Group.ExactCovering G ι), ∃ i j, i ≠ j ∧ (P.parts i).index = (P.parts j).index

What you must prove

import FormalConjectures.ErdosProblems.«274»
import TaskSupport

namespace Bounty

theorem target : fcTypeOfName% "Erdos274.herzog_schonheim" := by
  sorry

end Bounty

Pinned source: FormalConjectures/ErdosProblems/274.lean

Source type SHA-256
sha256:6e309ada359ca6dd3d37cc74ae7fe2288fce8c2c6332911204d420be9433a74b
Task id
fc-e923379e-erdos274-herzog-schonheim-f8e2742c92-formalized-v1
Task commitment
sha256:5b2776e52590679fdcf51f2d1755ae634a1e3f124aad4fcacee3552f94971b60

Something wrong with this formalization?

A statement that does not faithfully capture the original conjecture is the one real risk here, so we would rather hear about it early — before someone spends weeks on it.

Erdős Problem 274 · Conjectures.io