-
Notifications
You must be signed in to change notification settings - Fork 281
Add Erdős Problem 1167 (stepping-down partition relation) #3791
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Open
henrykmichalewski
wants to merge
3
commits into
google-deepmind:main
Choose a base branch
from
henrykmichalewski:add-problem-1167
base: main
Could not load branches
Branch not found: {{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
Open
Changes from 2 commits
Commits
Show all changes
3 commits
Select commit
Hold shift + click to select a range
File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,249 @@ | ||
| /- | ||
| 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 FormalConjectures.Util.ProblemImports | ||
|
|
||
| /-! | ||
| # Erdős Problem 1167 | ||
|
|
||
| **Verbatim statement (Erdős #1167, status O):** | ||
| > Let $r\geq 2$ be finite and $\lambda$ be an infinite cardinal. Let $\kappa_\alpha$ be cardinals for all $\alpha<\gamma$. Is it true that\[2^\lambda \to (\kappa_\alpha+1)_{\alpha<\gamma}^{r+1}\]implies\[\lambda \to (\kappa_\alpha)_{\alpha<\gamma}^{r}?\]Here $+$ means cardinal addition, so that $\kappa_\alpha+1=\kappa_\alpha$ if $\kappa_\alpha$ is infinite. | ||
|
|
||
| **Source:** https://www.erdosproblems.com/1167 | ||
|
|
||
| **Notes:** OPEN | ||
|
|
||
|
|
||
| *Reference:* [erdosproblems.com/1167](https://www.erdosproblems.com/1167) | ||
|
|
||
| ## Problem Statement | ||
|
|
||
| Let $r \geq 2$ be finite and $\lambda$ be an infinite cardinal. Let $\kappa_\alpha$ be cardinals | ||
| for all $\alpha < \gamma$. Is it true that | ||
| $$2^\lambda \to (\kappa_\alpha + 1)_{\alpha < \gamma}^{r+1}$$ | ||
| implies | ||
| $$\lambda \to (\kappa_\alpha)_{\alpha < \gamma}^r?$$ | ||
|
|
||
| Here the notation $\mu \to (\nu_\alpha)_{\alpha < \gamma}^s$ denotes the partition arrow: for | ||
| every coloring of $s$-element subsets of a set of cardinality $\mu$ using colors indexed by | ||
| $\alpha < \gamma$, there exists a monochromatic (homogeneous) subset of cardinality $\nu_\alpha$ | ||
| for some $\alpha < \gamma$. | ||
|
|
||
| This is a problem of Erdős, Hajnal, and Rado. | ||
|
|
||
| ## Overview | ||
|
|
||
| The central novelty in this file is `CardinalPartitionRel`: the general multicolor $r$-uniform | ||
| partition relation $\mu \to (\nu_\alpha)_{\alpha < \gamma}^r$ for infinite cardinals. | ||
|
|
||
| - `μ` (a `Cardinal`) is the size of the base set whose $r$-element subsets are colored. | ||
| - `r : ℕ` is the uniformity (edges are $r$-element finite sets). | ||
| - `γ` (an `Ordinal`) indexes the color classes. | ||
| - `ν : γ.ToType → Cardinal` assigns a target cardinality to each color. | ||
|
|
||
| The open question `erdos_1167` asks whether the $(r+1)$-uniform relation at $2^\lambda$ | ||
| (with targets $\kappa_\alpha + 1$) implies the $r$-uniform relation at $\lambda$ | ||
| (with targets $\kappa_\alpha$). | ||
|
|
||
| ## Formalization Notes | ||
|
|
||
| - **r-element subsets**: We use `{s : Finset α // s.card = r}` as the type of $r$-element | ||
| subsets of a type `α` with `#α = μ`. | ||
| - **Homogeneous set**: A set `H : Set α` is homogeneous for color `i` if every $r$-element | ||
| subset of `H` receives color `i`. | ||
| - **Cardinal successor**: For an infinite cardinal $\kappa$, $\kappa + 1 = \kappa$ in cardinal | ||
| arithmetic. The $+1$ in the antecedent is thus only nontrivial when $\kappa_\alpha$ is finite. | ||
| - We use `Ordinal.ToType` to turn the index ordinal `γ` into a type for the coloring function. | ||
| - We avoid the identifier `λ` (reserved in Lean 4) and write `lam` for the cardinal $\lambda$. | ||
| -/ | ||
|
|
||
| open Cardinal Ordinal | ||
|
|
||
| namespace Erdos1167 | ||
|
|
||
| universe u | ||
|
|
||
| /- ### The r-uniform cardinal partition relation -/ | ||
|
|
||
| /-- | ||
| `CardinalPartitionRel μ r γ ν` asserts the multicolor $r$-uniform partition relation | ||
| $$\mu \to (\nu_\alpha)_{\alpha < \gamma}^r.$$ | ||
|
|
||
| It states: for any type `A` with `#A = μ` and any coloring `col` of the $r$-element finite | ||
| subsets of `A` by `γ.ToType` (the colors indexed by ordinals less than `γ`), there exists: | ||
| - a color `i : γ.ToType`, and | ||
| - a subset `H : Set A` with `#H = ν i`, | ||
| such that every $r$-element subset of `H` receives color `i`. | ||
|
|
||
| When `γ = 2` and `r = 2` this reduces to the classical binary partition relation | ||
| $\mu \to (\nu_0, \nu_1)^2$. | ||
| -/ | ||
| def CardinalPartitionRel (μ : Cardinal.{u}) (r : ℕ) (γ : Ordinal.{u}) | ||
| (ν : γ.ToType → Cardinal.{u}) : Prop := | ||
| ∀ (A : Type u), #A = μ → | ||
| ∀ col : {s : Finset A // s.card = r} → γ.ToType, | ||
| ∃ (i : γ.ToType) (H : Set A), | ||
| #H = ν i ∧ | ||
| ∀ (s : Finset A) (hs : s.card = r), (↑s : Set A) ⊆ H → col ⟨s, hs⟩ = i | ||
|
|
||
| /- ### The main open problem -/ | ||
|
|
||
| /-- | ||
| **OPEN**: Let $r \geq 2$ be finite and $\lambda$ be an infinite cardinal. Let | ||
| $\kappa_\alpha$ be cardinals for all $\alpha < \gamma$. Is it true that | ||
| $$2^\lambda \to (\kappa_\alpha + 1)_{\alpha < \gamma}^{r+1} \implies | ||
| \lambda \to (\kappa_\alpha)_{\alpha < \gamma}^r?$$ | ||
|
|
||
| Here $\kappa_\alpha + 1$ denotes the cardinal successor (for finite $\kappa_\alpha$) or | ||
| $\kappa_\alpha$ itself (for infinite $\kappa_\alpha$, since $\kappa + 1 = \kappa$). | ||
|
|
||
| The relation $\mu \to (\nu_\alpha)_{\alpha < \gamma}^s$ is the $s$-uniform partition relation: | ||
| every coloring of $s$-element subsets of a $\mu$-sized set by $\gamma$ colors yields a | ||
| monochromatic set of size $\nu_\alpha$ for some $\alpha$. | ||
|
|
||
| The "stepping-down" goes from $(r+1)$-uniform hypergraphs on a set of size $2^\lambda$ to | ||
| $r$-uniform hypergraphs on a set of size $\lambda$. | ||
|
|
||
| *A problem of Erdős, Hajnal, and Rado.* | ||
|
|
||
| **Status**: OPEN. | ||
| -/ | ||
| @[category research open, AMS 5] | ||
| theorem erdos_1167 : | ||
| ∀ (r : ℕ), 2 ≤ r → | ||
| ∀ (lam : Cardinal.{u}), ℵ₀ ≤ lam → | ||
| ∀ (γ : Ordinal.{u}) (κ : γ.ToType → Cardinal.{u}), | ||
| -- Hypothesis: 2^lam → (κ_α + 1)_{α<γ}^{r+1} | ||
| CardinalPartitionRel ((2 : Cardinal.{u}) ^ lam) (r + 1) γ (fun α => κ α + 1) → | ||
| -- Conclusion: lam → (κ_α)_{α<γ}^r | ||
| CardinalPartitionRel lam r γ κ := by | ||
| sorry | ||
|
|
||
| /- ### Variants and related results -/ | ||
|
|
||
| namespace erdos_1167.variants | ||
|
|
||
| /-- | ||
| **Finite target case**: When all $\kappa_\alpha$ are finite, $\kappa_\alpha + 1$ in the | ||
| partition relation is the ordinary successor natural number. This is the "classical" instance | ||
| of the stepping-down problem: from $(r+1)$-uniform colorings on $2^\lambda$ to $r$-uniform | ||
| colorings on $\lambda$, with finite target sizes. | ||
|
|
||
| **Status**: OPEN (as a special case of `erdos_1167`). | ||
| -/ | ||
| @[category research open, AMS 5] | ||
| theorem finite_targets (r : ℕ) (hr : 2 ≤ r) (lam : Cardinal.{u}) (hlam : ℵ₀ ≤ lam) | ||
| (γ : Ordinal.{u}) (n : γ.ToType → ℕ) : | ||
| CardinalPartitionRel ((2 : Cardinal.{u}) ^ lam) (r + 1) γ | ||
| (fun α => (n α : Cardinal.{u}) + 1) → | ||
| CardinalPartitionRel lam r γ (fun α => (n α : Cardinal.{u})) := by | ||
| sorry | ||
|
|
||
| /-- | ||
| **Binary case**: The $\gamma = 2$ specialization, i.e. two color classes. | ||
|
|
||
| Is it true that $2^\lambda \to (\kappa_0 + 1, \kappa_1 + 1)^{r+1}$ implies | ||
| $\lambda \to (\kappa_0, \kappa_1)^r$ for all $r \geq 2$ and infinite $\lambda$? | ||
|
|
||
| **Status**: OPEN (as a special case of `erdos_1167`). | ||
| -/ | ||
| @[category research open, AMS 5] | ||
| theorem binary_colors (r : ℕ) (hr : 2 ≤ r) (lam κ₀ κ₁ : Cardinal.{u}) (hlam : ℵ₀ ≤ lam) : | ||
| -- γ = 2, colors indexed by Fin 2 | ||
| (∀ (A : Type u), #A = (2 : Cardinal.{u}) ^ lam → | ||
| ∀ col : {s : Finset A // s.card = r + 1} → Fin 2, | ||
| (∃ H : Set A, #H = κ₀ + 1 ∧ | ||
| ∀ (s : Finset A) (hs : s.card = r + 1), | ||
| (↑s : Set A) ⊆ H → col ⟨s, hs⟩ = 0) ∨ | ||
| (∃ H : Set A, #H = κ₁ + 1 ∧ | ||
| ∀ (s : Finset A) (hs : s.card = r + 1), | ||
| (↑s : Set A) ⊆ H → col ⟨s, hs⟩ = 1)) → | ||
| (∀ (A : Type u), #A = lam → | ||
| ∀ col : {s : Finset A // s.card = r} → Fin 2, | ||
| (∃ H : Set A, #H = κ₀ ∧ | ||
| ∀ (s : Finset A) (hs : s.card = r), | ||
| (↑s : Set A) ⊆ H → col ⟨s, hs⟩ = 0) ∨ | ||
| (∃ H : Set A, #H = κ₁ ∧ | ||
| ∀ (s : Finset A) (hs : s.card = r), | ||
| (↑s : Set A) ⊆ H → col ⟨s, hs⟩ = 1)) := by | ||
| sorry | ||
|
|
||
| /-- | ||
| **Infinite target case**: When all $\kappa_\alpha \geq \aleph_0$ are infinite, the cardinal | ||
| successor satisfies $\kappa_\alpha + 1 = \kappa_\alpha$ in cardinal arithmetic, so the | ||
| hypothesis simplifies to | ||
| $$2^\lambda \to (\kappa_\alpha)_{\alpha < \gamma}^{r+1}$$ | ||
| implying | ||
| $$\lambda \to (\kappa_\alpha)_{\alpha < \gamma}^r.$$ | ||
|
|
||
| The $+1$ disappears for infinite targets, making this a "pure" stepping-down lemma. | ||
|
|
||
| **Status**: OPEN (as a special case of `erdos_1167`). | ||
| -/ | ||
| @[category research open, AMS 5] | ||
| theorem infinite_targets (r : ℕ) (hr : 2 ≤ r) (lam : Cardinal.{u}) (hlam : ℵ₀ ≤ lam) | ||
| (γ : Ordinal.{u}) (κ : γ.ToType → Cardinal.{u}) (hκ : ∀ i, ℵ₀ ≤ κ i) : | ||
| -- For infinite κ_α, κ_α + 1 = κ_α (so hypothesis and conclusion share the same targets) | ||
| CardinalPartitionRel ((2 : Cardinal.{u}) ^ lam) (r + 1) γ κ → | ||
| CardinalPartitionRel lam r γ κ := by | ||
| sorry | ||
|
|
||
| /-- | ||
| **r = 2 case**: The stepping-down from 3-uniform to 2-uniform (pairs) partition relations. | ||
|
|
||
| Is it true that $2^\lambda \to (\kappa_\alpha + 1)_{\alpha < \gamma}^3$ implies | ||
| $\lambda \to (\kappa_\alpha)_{\alpha < \gamma}^2$? | ||
|
|
||
| This is the case $r = 2$ of `erdos_1167` and is a generalization of the classical | ||
| Erdős–Rado stepping-up/stepping-down theorems for pairs. | ||
|
|
||
| **Status**: OPEN (as a special case of `erdos_1167`). | ||
| -/ | ||
| @[category research open, AMS 5] | ||
| theorem r_eq_two (lam : Cardinal.{u}) (hlam : ℵ₀ ≤ lam) | ||
| (γ : Ordinal.{u}) (κ : γ.ToType → Cardinal.{u}) : | ||
| CardinalPartitionRel ((2 : Cardinal.{u}) ^ lam) 3 γ (fun α => κ α + 1) → | ||
| CardinalPartitionRel lam 2 γ κ := by | ||
| sorry | ||
|
|
||
| end erdos_1167.variants | ||
|
|
||
| /- ### Sanity checks -/ | ||
|
|
||
| /-- | ||
| **Sanity check**: The hypothesis `ℵ₀ ≤ lam` is non-vacuous: $\aleph_0$ is an infinite cardinal. | ||
| -/ | ||
| @[category test, AMS 5] | ||
| example : ℵ₀ ≤ ℵ₀ := le_refl _ | ||
|
|
||
| /-- | ||
| **Sanity check**: $2^{\aleph_0} > \aleph_0$, so the power $2^\lambda$ is strictly larger than | ||
| $\lambda$ for $\lambda = \aleph_0$. This confirms the two base sets in the stepping-down | ||
| implication are genuinely of different sizes. | ||
| -/ | ||
| @[category test, AMS 5] | ||
| example : ℵ₀ < (2 : Cardinal) ^ ℵ₀ := Cardinal.cantor ℵ₀ | ||
|
|
||
| /-- | ||
| **Sanity check**: Every 0-element finset is empty. This confirms `Finset.card = 0` is | ||
| non-trivial and that the condition `s.card = r` in `CardinalPartitionRel` correctly captures | ||
| the notion of an $r$-element subset. | ||
| -/ | ||
| @[category test, AMS 5] | ||
| example (A : Type u) (s : Finset A) (hs : s.card = 0) : s = ∅ := | ||
| Finset.card_eq_zero.mp hs | ||
|
|
||
| end Erdos1167 | ||
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
none of these are really useful: they are just some theorems in mathlib that don't need repeating here -- those definitions are sufficiently tested in mathlib