Weighted Zero-Sum Subsequence Theory

• Weighted zero-sum subsequences are sequences of group elements with assigned weights that sum to zero, forming a fundamental structure in combinatorial number theory.
• They reveal deep structural properties of finite abelian groups, with key invariants such as the weighted Davenport constant and extremal sequence constructions.
• Recent developments, especially in the plus-minus weighted case, integrate factorization theory, commutative algebra, and additive combinatorics to solve open problems.

A weighted zero-sum subsequence is a fundamental structure in additive and combinatorial number theory, where one seeks subsequences of group elements such that their weighted sum, under prescribed sets of coefficients (“weights”), vanishes in the group. The precise nature of “weighting,” the types of groups considered, and the corresponding extremal invariants (e.g., Davenport constants, extremal sequences, structural monoid properties) yield a rich and highly developed theory, especially in the context of finite abelian groups. The plus-minus weighted zero-sum case (where weights are restricted to $\{\pm1\}$) has recently received a comprehensive algebraic and arithmetic treatment, synthesizing techniques from factorization theory, commutative algebra, and additive combinatorics (Fabsits et al., 2023, Geroldinger et al., 2022).

1. Definitions and Core Principles

Let $G$ be an (additively written) abelian group. The free abelian monoid $F(G)$ consists of finite sequences $S = g_1 \cdot \ldots \cdot g_\ell$ with terms $g_i \in G$; the monoid operation is concatenation. A sequence $S \in F(G)$ is called a plus-minus weighted zero-sum sequence if there are signs $\varepsilon_i \in \{-1,1\}$ such that

$$\varepsilon_1 g_1 + \cdots + \varepsilon_\ell g_\ell = 0 \quad \text{in } G.$$

The collection of all plus-minus weighted zero-sum sequences forms the submonoid $\mathcal{B}_{\pm}(G) \subset F(G)$, equipped with concatenation as its operation and the empty sequence as its identity.

More generally, for a given set of group automorphisms $\Gamma \subseteq \operatorname{End}(G)$ (“weight-set”), a $\Gamma$-weighted zero-sum sequence is $S = g_1 \cdot \ldots \cdot g_\ell \in F(G)$ for which there exist $\gamma_i \in \Gamma$ with

$\gamma_1(g_1) + \cdots + \gamma_\ell(g_\ell) = 0.$

The monoid of all such sequences is denoted $\mathcal{B}_\Gamma(G)$ (Geroldinger et al., 2022).

2. Algebraic and Structural Properties

The monoid $\mathcal{B}_{\pm}(G)$ admits a rich ideal-theoretic and algebraic theory:

• Mori property: $\mathcal{B}_{\pm}(G)$ is a Mori monoid (i.e., satisfies ACC on divisorial/v-ideals) if and only if $2G$ is finite, equivalently $G = G_1 \oplus G_2$ with $G_1$ an elementary $2$-group and $G_2$ finite.
• Krull property: $\mathcal{B}_{\pm}(G)$ is a Krull monoid if and only if $G$ is an elementary $2$-group. Hence, for almost all $G$ beyond the trivial ($C_2^{\oplus r}$) cases, the plus-minus weighted zero-sum monoid is not Krull.
• Finite generation (C-monoid): $\mathcal{B}_{\pm}(G)$ is finitely generated if and only if $G$ is finite. In this case, its atoms (irreducibles) are exactly the minimal plus-minus weighted zero-sum sequences; these are necessarily finite in number.

A parallel theory holds for more general $\Gamma$; the monoids $\mathcal{B}_\Gamma(G)$ are always reduced, finitely generated, and cancellative for finite $G$, with factorization invariants controlled by the structure of $G$ and the weight-set $\Gamma$ (Geroldinger et al., 2022). Seminormality can be characterized in terms of group exponent ($\operatorname{exp}(G) \mid 4$ for $\mathcal{B}_{\pm}(G)$).

3. Invariants and Extremal Constants

The study of weighted zero-sum subsequences centers around several extremal invariants:

• Weighted Davenport constant $D_{A}(G)$: Minimal $\ell$ so every sequence of length $\ell$ over $G$ contains a nonempty $A$-weighted zero-sum subsequence, where $A$ is the weight-set (e.g., $A = \{\pm1\}$ in the plus-minus case).
• Counting function $N_{A,0}(S)$: Number of nonempty subsequences of $S$ whose $A$-weighted sum is zero. For $A = \{1,2,\ldots,n-1\}$ and $G$ of exponent $n$, the bound $N_{A,0}(S) \geq 2^{|S|-D_A(G)+1}$ holds, and is sharp for extremal $S$ (Lemos et al., 2015, Lemos et al., 2018, Lemos et al., 2022).
• Consecutive weighted zero-sum constant $C_A(n)$: Minimal $k$ so every sequence of length $k$ in $\mathbb{Z}_n$ has a (consecutive) $A$-weighted zero-sum subsequence. For $A = \{\pm1\}$ in cyclic or direct sum settings, explicit recursive/closed formulas are obtained, e.g., $C_{\{\pm1\}}(n) = n$ for $n$ odd (Mondal et al., 2021).
• Unit-weighted and higher residue-structure constants: For $A = U(n)$, $A = U(n)^2$, $A = U(n)^3$ and $n$ odd, explicit closed forms for $C_{A}(n)$ and $D_A(n)$ are known and admit recursive constructions depending on prime decomposition and residue degree (Mondal et al., 2021, Mondal et al., 2022, Paul et al., 2022).

4. Isomorphism and Characterization Problems

A key structural question is the isomorphism problem: for which $G_1, G_2$ does $$\mathcal{B}_{\pm}(G_1) \cong \mathcal{B}_{\pm}(G_2)$$ as monoids imply $G_1 \cong G_2$ as groups? For finite direct sums of cyclic groups, the answer is affirmative: an isomorphism of monoids necessarily induces a group isomorphism (Fabsits et al., 2023). The proof extracts invariants from factorization structure, notably through atom-squares.

A related question is the characterization problem: does equality of systems of lengths $$\mathcal{L}(\mathcal{B}_{\pm}(G_1)) = \mathcal{L}(\mathcal{B}_{\pm}(G_2))$$ imply $G_1 \cong G_2$? This is resolved positively for cyclic groups of odd order $n \geq 5$ (determined by factorization lengths) and for various small groups, with the exception of certain index-two phenomena in composite order cases (Fabsits et al., 2023, Geroldinger et al., 2022).

$G$	$\mathcal{B}_{\pm}(G)$ property	Main invariant/control
Elementary $2$-group	Krull monoid	Krull class group/C-monoid
$G$ finite	Finitely generated	Atoms enumerate all minimal $\pm 1$ sums
$G = C_n$, $n$ odd $\geq 5$	Sets of lengths characterize $G$	$\mathcal{L}(\mathcal{B}_{\pm}(C_n))$
$G = \mathbb{Z}$	Not Mori	Infinite $2G$

5. Explicit Constructions and Extremal Sequences

For each weighted zero-sum invariant, sharp constructions realize the critical threshold:

• For $D_A(G)$ (e.g., $A = \{\pm1\}$, $G = C_3^r$), extremal sequences avoid nontrivial $A$-weighted zero-sum subsequences up to the critical length, by explicit combinatorial or linear-algebraic constructions (Godinho et al., 2011).
• For $C_{U(n)}(n)$, extremal sequences for the unit-weighted consecutive constant are constructed recursively: for $n=2^k$, a C-extremal sequence contains exactly one odd term (at the midpoint) and all others even, with recursively C-extremal left and right halves after division by 2 (Mondal et al., 2022).
• For non-principal weight-sets (e.g., cubes, squares), extremal sequences combine prime-power and CRT-local constructions, with the critical length determined by the corresponding local invariants (Sarkar, 2022, Paul et al., 2022).

Minimal zero-sum sequences and the structure of the extremals for enumeration bounds (e.g., sequences achieving $N_{A,0}(S) = 2^{|S|-D_A(G)+1}$) have been systematically classified for finite cyclic and direct product groups (Lemos et al., 2018, Lemos et al., 2015, Lemos et al., 2022).

6. Connections to Factorization Theory and Arithmetic Applications

The investigation of monoids of weighted zero-sum sequences is tightly bound to non-unique factorization theory. A monoid $H$ is atomic if every nonunit factors into irreducibles ("atoms"), and much of the arithmetic structure (lengths of factorizations, catenary degrees, $\omega$-invariants, union and distance sets) is determined by the configuration of minimal zero-sum sequences:

• For norm monoids in Galois number fields, transfer homomorphisms to monoids of weighted zero-sum sequences show that the factorization theory of number rings is governed by the corresponding zero-sum monoid (Geroldinger et al., 2022).
• Sets of lengths in weighted monoids often form intervals or unions with endpoints explicitly described; for $G = C_p$ and plus-minus weighting, the set of distances is $[1, p-2]$ and major invariants (catenary degree, $\omega$-invariant) are all equal to $p$ (Geroldinger et al., 2022).
• Open questions remain regarding realization of possible length sets, monotone catenary degree comparison between monoids, and the full scope of the invertibility of the length-system-characterization mapping.

The structure and arithmetic of $\mathcal{B}_{\pm}(G)$ encode both combinatorial and monoidal complexity, with deep links to classical invariants and modern factorization-theoretic techniques (Fabsits et al., 2023, Geroldinger et al., 2022).

7. Open Problems and Research Directions

Active areas of inquiry include:

• Extension to infinite abelian groups: The universal zero-sum invariant framework generalizes these problems and admits interpretations in infinite settings, with covering theory (Neumann) linking finiteness of weighted Davenport constants to group structure (Wang, 2022).
• Classification for general weight-sets: While much is known for $A = \{\pm1\}, U(n), U(n)^k$, further classes (arbitrary subgroups, "Jacobi symbol" weights, polynomial images) remain to be fully characterized with respect to invariants such as $D_A(G)$ and $C_A(G)$ (Mondal et al., 2022).
• Factorization set-determination: To what extent does the system of sets of lengths or distance sets determine the group up to isomorphism for $\mathcal{B}_{\pm}(G)$?
• Structural arithmetic in monoids of higher weights: Determining Krull, seminormal, and root-closed properties in wider classes of weighted zero-sum monoids, and understanding their role as transfer targets from number-theoretic monoids.

Continued research is likely to focus on the interplay between higher-order weightings, noncyclic groups, modular parameters, and arithmetic invariants.

---

References:

• (Fabsits et al., 2023): On Monoids of plus-minus weighted Zero-Sum Sequences: The Isomorphism Problem and the Characterization Problem
• (Geroldinger et al., 2022): On monoids of weighted zero-sum sequences and applications to norm monoids in Galois number fields and binary quadratic forms
• (Mondal et al., 2021): Extremal sequences for a weighted zero-sum constant
• (Mondal et al., 2021): On a different weighted zero-sum constant
• (Lemos et al., 2015): On the number of weighted subsequences with zero-sum in a finite abelian group
• (Lemos et al., 2018): On the number of fully weighted zero-sum subsequences
• (Lemos et al., 2022): On the Number of Weighted Zero-sum Subsequences
• (Mondal et al., 2021): Zero-sum constants related to the Jacobi symbol
• (Sarkar, 2022): Generalization of some weighted zero-sum theorems and related Extremal sequence
• (Wang, 2022): The universal zero-sum invariant and weighted zero-sum for infinite abelian groups
• (Godinho et al., 2011): Weighted Zero-Sum Problems Over $C_3^r$
• (Mondal et al., 2022): On unit-weighted zero-sum constants of $\mathbb Z_n$