One color for every finite subset sum and product

← All digests

Family 164 · Combinatorics
Importance 5/5 · Provisional editorial rating
Sources consulted 2026-10-11

1. The question

Color every positive integer using finitely many colors. Can one find distinct numbers so that adding any nonempty selection and multiplying any nonempty selection always gives the same color?

For two chosen numbers $x,y$, this asks that $x,y,x+y,xy$ all share a color. For three, it asks simultaneously about the individual numbers, the three pairwise sums, the total sum, the three pairwise products and the total product. As the desired set grows, the number of constraints grows exponentially.

The manuscript in family 164 claims a positive answer for every prescribed finite size, resolving Hindman’s finite sums and products conjecture over the positive integers. Its statement also allows the selected numbers to grow very rapidly, ensuring that the requested expressions are genuinely distinct except where a singleton sum and product necessarily agree.

The finiteness qualification is essential: an infinite simultaneous version is known to fail.

2. Background and essential definitions

For a finite set $A$ of positive integers, define

$$\operatorname{FS}(A)= \left\{\sum_{a\in B}a:\varnothing\ne B\subseteq A\right\},$$

and

$$\operatorname{FP}(A)= \left\{\prod_{a\in B}a:\varnothing\ne B\subseteq A\right\}.$$

The initials stand for finite sums and finite products. Each selected element is used at most once in an expression. Empty selections are excluded, so neither the empty sum zero nor the empty product one is imposed by definition. Singleton selections are included, so $A$ itself must have the common color.

A coloring is any function $\chi:\mathbb N\to\{1,\ldots,r\}$, where $r$ is finite. There is no assumption that colors have a simple formula or positive density. A pattern is called partition regular if some monochromatic instance occurs in every finite coloring.

For illustration, if $A=\{2,5\}$, the required union is $\{2,5,7,10\}$. For $A=\{2,5,11\}$, both 18, the total sum, and 110, the total product, must share the color of all the other required expressions. The theorem chooses a suitable set for the coloring; it does not say these illustrative sets work in every coloring.

Addition and multiplication respond differently to scaling. Replacing every element by $Da$ multiplies a subset sum by $D$, but a product of $j$ elements by $D^j$. An arbitrary coloring need not respect any of these changes. This explains why a rational configuration cannot simply be turned into an integer one by clearing denominators.

3. Development history and previous partial results

Schur’s theorem guarantees a monochromatic triple $x,y,x+y$. The finite sums theorem of Folkman–Rado–Sanders extends additive monochromaticity to all nonempty subset sums of an arbitrarily large finite set. Hindman’s 1974 theorem goes further, producing an infinite sequence whose finite sums all have one color.

Multiplicative counterparts follow by applying an additive theorem to the coloring of powers of a fixed integer: adding exponents multiplies the original numbers. However, this does not make the same chosen numbers work simultaneously for addition and multiplication.

Hindman formulated the finite simultaneous question in his 1979 work. His counterexample work shows that an infinite simultaneous assertion cannot hold in general. Arbitrarily large finite witnesses therefore represent a meaningful endpoint, not merely an intermediate version of an automatic infinite theorem.

Important partial patterns were established. Moreira’s 2017 theorem gives monochromatic triples $x,x+y,xy$ in arbitrary finite colorings, within a broader polynomial framework. Notice that the color of $y$ is absent from this triple. Results for two colors, including Bowen’s work, address the four-expression pattern and further related configurations.

Over the rational numbers, Alweiss proved the full finite sums and products statement. The integer problem retains an arithmetic obstacle because the compensating scalings available over a field need not preserve integral variables. The new manuscript adapts parts of that finite selection and alignment strategy while supplying the required integer-scale control.

4. What the new manuscripts claim

For every finite coloring and every integer $m\ge1$, M1’s Theorem 1.1 asserts distinct positive integers $a_1<\cdots<a_m$ and a color $c$ such that

$$\chi\!\left(\sum_{j\in J}a_j\right) =\chi\!\left(\prod_{j\in J}a_j\right)=c \qquad(\varnothing\ne J\subseteq\{1,\ldots,m\}).$$

The theorem has an additional separation property. Given real parameters $R\ge2$ and $D\ge1$, it can choose $a_1>R$ and, for each $2\le d\le m$,

$$a_d>R\left(\sum_{j<d}a_j+\prod_{j<d}a_j\right)^D.$$

Each next element can thus exceed a prescribed power of the combined earlier sum and product. This prevents the result from depending on coincidences among expressions or repeated use of a small special value.

Corollary 1.2 states that the set can be chosen with

$$|\operatorname{FS}(A)|=|\operatorname{FP}(A)|=2^m-1, \qquad \operatorname{FS}(A)\cap\operatorname{FP}(A)=A.$$

Consequently the monochromatic union has $2(2^m-1)-m$ distinct elements. The shared singleton expressions account for the subtraction.

The parameters are qualitative. The source does not provide a useful numerical upper bound for the smallest possible witness or a practical search algorithm for an arbitrary coloring.

5. Why the advance matters

Importance: 5/5. The simultaneous finite sums and products conjecture is a central problem about the interaction of the two arithmetic operations. Separate additive and multiplicative regularity is much easier to obtain than one configuration respecting both at once.

The claimed result would establish arbitrarily large finite patterns even though the corresponding infinite behavior can be obstructed. That sharply illustrates how finite Ramsey phenomena can coexist with negative infinite results.

The integer setting adds significance beyond the known rational theorem. Denominator and divisibility constraints are part of the problem’s structure, and solving them while keeping a single color requires more than a formal change of domain.

The separation clause is also informative: the monochromatic pattern can be forced far out among large, widely separated numbers, with all its intended sums and products distinct.

6. The main idea

Secure products, then coordinate the sums

The manuscript organizes raw integer variables into ordered blocks with widely separated scales. Finite Ramsey and finite-sums selection choose block products whose nonempty products already share a color. This supplies the multiplicative part of the target.

The remaining challenge is additive. Every sum of selected block products must retain that same color, including sums involving blocks from very different scales. Choosing a new additive configuration independently would lose the multiplicative agreement already established.

The proof separates this challenge into two principles, called Prediction and Alignment. Prediction replaces irregular color indicators by controlled models inside the relevant counting expressions. Alignment chooses changes of scale that preserve positive predictions at all required shifted locations.

Model color indicators with the correct sampling law

The indicator of a color is the function taking value one on that class and zero elsewhere. The source approximates these indicators by piecewise nilsequences, structured sequences generalizing functions of polynomial phases. Their role is to make certain translations predictable enough to coordinate.

A block product is not distributed like a uniformly sampled integer of comparable size. Earlier factors force divisibility conditions, so the counting model includes nonnegative divisor weights. The same weights must remain attached to the related sums. Otherwise, an approximation valid under one sampling distribution would not justify the expressions actually being selected.

The source uses prime substitutions and repeated Cauchy–Schwarz estimates to control the discrepancy while retaining the already imposed product-color conditions. It also builds models on nested additive scales, since local counting and the later alignment procedure need information over different interval sizes.

Align without losing integrality

The rational strategy permits compensating scale changes that keep a target product fixed. In the integer problem, these changes can create nonintegral shifts in a raw variable. M1 first makes an integer residue correction, then represents the remaining displacement through transformations of the structured model.

Several models must respond coherently to the same change. The manuscript supplies a lifting step that puts their transformations on a common state space. A finite recurrence theorem for the resulting nilpotent action then yields an alignment with positive conditional probability.

The important quantitative feature is qualitative uniformity: the lower bound for successful alignment is independent of the full complexity of the models. This lets the approximation errors be chosen small enough afterward, so a positive model prediction leads to an actual colored configuration.

A final finite selection assembles the blocks, restores integral variables and enforces the rapid separation. This last step produces one set satisfying all subset-sum and subset-product conditions together.

7. Scope, limitations, and open questions

The theorem concerns every prescribed finite size. It does not provide one infinite set whose sums and products all share a color; Hindman’s counterexamples explain why that stronger assertion is unavailable.

The expressions use distinct selected elements, with no repetitions inside a sum or product. The result therefore does not automatically force every polynomial expression, every power of an element, or every expression mixing addition and multiplication in arbitrary arrangements.

No density assumption is required of a particular color class, but the theorem also does not identify in advance which color succeeds. The successful color may depend on the requested size and separation parameters.

The source’s qualitative construction leaves quantitative questions: how large a finite interval suffices for a given number of colors and desired set size, how efficiently a witness can be found, and which further mixed arithmetic patterns can be added. These require information beyond the stated existence theorem.

8. Sources and further reading

M1’s Theorem 1.1 gives the simultaneous pattern and separation, and Corollary 1.2 records distinctness of the resulting expressions. Section 2 explains the finite combinatorial deduction. Sections 3–5 develop Prediction, while Sections 6–8 develop Alignment and the return to integers.

The primary references linked in Section 3 distinguish the additive theorem, the negative infinite simultaneous result, the mixed-triple advance and the rational finite theorem. Wikipedia’s Hindman’s theorem is optional background on the additive starting point.

Source manuscripts

• M1. OpenAI, Monochromatic finite sums and products in the positive integers (September 23, 2026).

Sources consulted on 11 October 2026 at repository revision fd4aeeb. Download the manuscript citations.