🔢 mathematics

$N$ -Koszul algebras of finite global dimension for $N\geq 3$

This paper proves that $N$ -Koszul AS regular algebras of finite global dimension must be the known 3-Koszul examples of global dimension 3, provided their Hilbert series pole order at $t=1$ exceeds a specific threshold or, more generally, if they possess the Hilbert series of weighted polynomial rings and their GK dimension equals their global dimension.

Original authors: So Nakamura

Published 2026-08-24

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: So Nakamura

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

In the vast landscape of mathematics, there is a branch dedicated to the study of shapes and spaces that are built from numbers and equations rather than clay or stone. This field, known as algebraic geometry, often begins with the simplest and most familiar building blocks: polynomial rings. These are mathematical structures that behave like the coordinate grids used to map the physical world, allowing mathematicians to describe curves, surfaces, and higher-dimensional forms with precision. For centuries, mathematicians have understood how these structures work when they follow the standard rules of arithmetic, where the order of multiplication does not matter. However, a more mysterious frontier exists where the order of operations does change the outcome, creating "noncommutative" spaces. In this realm, the familiar rules of geometry break down, and the structures that once served as reliable maps can become chaotic or even cease to exist in a meaningful way. To navigate this strange territory, researchers look for special types of algebras that retain enough order to be studied, much like finding islands of stability in a turbulent sea. Among these, a specific class known as Artin-Schelter regular algebras has long been suspected to be the most promising candidates for these noncommutative maps, but a critical question has remained unanswered: do these structures exist in forms that are more complex than the few simple examples already discovered?

A recent paper by So Nakamura tackles this question by investigating a specific family of these mathematical objects called N-Koszul algebras. The term "N-Koszul" refers to a property that describes how these algebras are constructed and how their internal components relate to one another, with the number N indicating the complexity of the construction rules. While mathematicians have known for some time that these structures exist when the complexity is low, specifically when N equals three, there has been no confirmed example of such a structure for any higher value of N. The mathematical community has long suspected that no such higher-complexity examples exist, but suspicion is not proof. Nakamura's work moves beyond conjecture to provide a rigorous demonstration that, under specific and widely accepted assumptions regarding the algebra's growth patterns, these more complex structures simply cannot exist. The paper does not merely suggest that they are unlikely; it proves that if such an algebra were to exist and satisfy conditions where its Hilbert series matches that of a weighted polynomial ring, its GK dimension coincides with its global dimension, and the order of the pole of its Hilbert series at t = 1 is greater than a specific threshold relative to the global dimension, it would have to violate fundamental mathematical constraints, leading to a logical contradiction.

The investigation begins by examining the "fingerprint" of these algebras, a mathematical signature known as the Hilbert series. This series acts like a growth chart, tracking how the size of the algebra expands as one moves through its layers. For the algebras in question, this growth is expected to follow a very specific pattern, similar to how a weighted polynomial ring expands. The author also considers the "global dimension," a measure of how many steps it takes to resolve the algebra's internal conflicts, and the "GK dimension," which measures the overall rate of growth. The core of the proof involves a delicate balancing act between these different measures of size and complexity, specifically requiring that the order of the pole of the Hilbert series at t = 1 be greater than a specific threshold relative to the global dimension. By assuming that an algebra with a complexity level N greater than three exists and satisfies these growth conditions, the author constructs a mathematical scenario that must hold true if such an object were real. This scenario involves a specific polynomial equation that describes the algebra's structure.

The proof proceeds by testing this scenario against the known rules of number theory and algebra. The author demonstrates that for any complexity level N that is a prime number greater than three, the required polynomial equation leads to an impossible situation. The argument relies on analyzing the roots of the polynomial and the way they interact with the growth rates of the algebra. Through a series of logical deductions, the paper shows that the only way to satisfy all the necessary conditions is if the complexity level N is exactly three. Any attempt to push the complexity higher, to N equals five or seven or beyond, causes the mathematical structure to collapse under its own weight, provided the algebra adheres to the standard growth patterns expected of such objects. The paper explicitly rules out the existence of these higher-complexity algebras under these specific assumptions, establishing that the known examples of complexity three are the only ones that can exist within this specific framework, assuming the conditions hold.

The significance of this result lies in its ability to close a long-standing chapter in the classification of noncommutative spaces, albeit conditionally. By proving that no other examples exist under the stated assumptions, the paper effectively maps the boundaries of this particular mathematical territory. It confirms that the universe of these specific algebras is far more limited than previously hoped, containing only the structures that have already been identified, provided the conjectured properties of AS regular algebras (such as having a Hilbert series of a weighted polynomial ring and matching GK and global dimensions) are true. This does not diminish the importance of the field; rather, it clarifies the landscape, allowing researchers to focus their efforts on understanding the properties of the existing structures rather than searching for ghosts in the machine. The work stands as a definitive proof that the search for these specific higher-complexity algebras has reached its natural conclusion within the scope of these assumptions, leaving the mathematical community with a complete and verified list of possibilities for this class of objects under those conditions.

Technical Summary: N-Koszul Algebras of Finite Global Dimension for $N \ge 3$

Problem Statement
The paper addresses the classification of $N$ -Koszul Artin-Schelter (AS) regular algebras for $N \ge 3$ . While AS regular algebras of global dimension 3 are well-understood (classified by Artin, Tate, and Van den Bergh), the existence of such algebras for $N \ge 3$ with finite global dimension other than the known 3-Koszul examples remains an open question.

Recent work by Kabbaj established that if an $N$ -Koszul algebra $A$ of finite global dimension $d$ has a Hilbert series corresponding to a weighted polynomial ring and its trivial module $K$ admits a finite free resolution, then $N$ must be a prime number and $d$ must be odd ( $d = 2k+1$ ). Furthermore, Kabbaj conjectured that under these conditions, the only solutions are the known 3-Koszul AS regular algebras of global dimension 3 (i.e., $N=3$ and $k=1$ ). This paper aims to prove this conjecture under a specific constraint regarding the growth of the algebra.

Methodology
The author employs a combination of homological algebra, combinatorial analysis of Hilbert series, and number-theoretic arguments. The core methodology involves:

Characteristic Polynomial Analysis: Utilizing the assumption that the trivial module $K$ has a finite free resolution, the Hilbert series $h_A(t)$ is related to the characteristic polynomial $p(t)$ of $K$ via $h_A(t) = 1/p(t)$ .
Structural Constraints: The polynomial $p(t)$ $p (t)$ is analyzed in two forms:
- As a product of terms $(1-t^i)^{n_i}$ (derived from the weighted polynomial ring hypothesis).
- As a specific symmetric polynomial $q(t) - t^{kN+1}q(1/t)$ derived from the $N$ -Koszul resolution structure, where $q(t) = \sum_{i=0}^k \alpha_i t^{iN}$ .
Power Sum Sequences: The author defines a sequence $\{\pi_i\}$ representing the power sums of the roots of $p(t)$ . Using logarithmic differentiation and Bell polynomials, explicit formulas for $\pi_i$ are derived in terms of the coefficients $\alpha_i$ and the parameter $s = n_1$ .
Inequality Bounding: A critical step involves bounding the integer $k$ (related to the global dimension) in terms of the exponents $n_i$ and the prime $N$ . This relies on the assumption that the order of the pole of $h_A(t)$ at $t=1$ (which equals $\sum n_i$ ) is sufficiently large relative to $k$ .
Number-Theoretic Elimination: The proof utilizes Nagura's generalization of Bertrand's postulate to show that for sufficiently large primes $N$ , there exists a prime $l$ within a specific range that forces a contradiction in the divisibility properties of the sequence $\pi_i$ .

Key Contributions and Results

Main Theorem (Theorem 0.6): The paper proves that if $N$ is an odd prime, $q(t) = \sum_{i=0}^k \alpha_i t^{iN}$ has positive integer coefficients, and the associated polynomial $p(t) = q(t) - t^{kN+1}q(1/t)$ factors as $\prod (1-t^i)^{n_i}$ with $\sum n_i > \frac{21}{11}k + 1$ , then it must be that $N=3$ , $k=1$ , and $p(t) = (1-t)^2(1-t^2)$ .
Corollary on AS Regular Algebras (Corollary 0.7): The author proves that any $N$ $N$ -Koszul AS regular algebra $A$ $A$ of finite global dimension $d$ $d$ is necessarily a 3-Koszul AS regular algebra of global dimension 3, provided:
1. $A$ has a Hilbert series of a weighted polynomial ring.
2. The order of the pole of its Hilbert series at $t=1$ is greater than $\frac{21d+1}{22}$ .
Corollary on GK Dimension (Corollary 0.8): Under the standard conjectures that AS regular algebras have a Hilbert series of a weighted polynomial ring and that their GK dimension equals their global dimension, the paper concludes that any $N$ -Koszul AS regular algebra for $N \ge 3$ must be a 3-Koszul algebra of global dimension 3.

Significance and Claims
The paper claims to significantly narrow the search space for non-commutative analogs of polynomial rings. By establishing that the "gap" between the global dimension and the GK dimension (represented by the order of the pole) cannot be arbitrarily small without forcing the algebra to collapse to the known 3-Koszul case, the work provides strong evidence for Kabbaj's conjecture.

The author notes that the coefficient $\frac{21}{11}$ in the main theorem is derived from specific inequalities in the proof (specifically inequality $(**)$ in Corollary 2.6) and acknowledges that it might be improvable with more extensive computation. The paper explicitly poses the question of whether the result holds without the assumption on the order of the pole (Assumption 2), noting that a positive answer would fully resolve Kabbaj's conjecture.

The work does not claim to construct new algebras but rather to prove the non-existence of $N$ -Koszul AS regular algebras for $N \ge 3$ (other than the 3-Koszul case) under the stated growth conditions, thereby reinforcing the uniqueness of the dimension 3 classification in this context.

Technical Summary: N-Koszul Algebras of Finite Global Dimension for N≥3N \ge 3N≥3

More like this

Technical Summary: N-Koszul Algebras of Finite Global Dimension for $N \ge 3$