Group Cohomology Classifications

• Group Cohomology Classifications are frameworks that organize algebraic and topological invariants using cohomology, spectral sequences, and tensor-triangular geometry.
• They stratify derived categories via localizing subcategories and nilpotence indices, enabling precise analysis of representations and group extensions.
• The approach integrates enriched data like the Bockstein spectral sequence, supporting comprehensive studies in both finite and infinite-dimensional group settings.

Group cohomology classifications organize and stratify the structures arising from the cohomology of algebraic and topological groups, with particular focus on representations, extensions, and derived categories. The classification problem interacts with triangulated categories, tensor-triangular geometry, spectral sequences, and the interplay between algebraic and topological invariants. The modern landscape includes the localizing subcategory classification for cochains on classifying spaces, nilpotence invariants in cohomology rings, spectral methods for unipotent and Kac–Moody groups, and refinements arising from enriched and equivariant perspectives.

1. Localizing Subcategories and Stratification

The derived category of cochains on the classifying space of a finite group, $D(C^*(BG;k))$, is a compactly generated triangulated category equipped with an action of the group cohomology ring $R = H^*(G, k)$. The core classification theorem establishes a bijection between localizing subcategories of $D(C^*(BG; k))$ and subsets of the set of homogeneous prime ideals $\operatorname{Spec}^h R$ of the cohomology ring. Explicitly, for localizing subcategories $\mathcal{S} \subseteq D(A)$, the correspondence is given by

$$\mathcal{S} \longleftrightarrow \bigcup_{X\in\mathcal{S}} \operatorname{supp}_R(X)$$

and its inverse

$$U \subseteq \operatorname{Spec}^h R \mapsto \{ X \in D(A) \mid \operatorname{supp}_R(X) \subseteq U \}.$$

Support is determined via local cohomology functors $\Gamma_\mathfrak{p}$, with $$\operatorname{supp}_R(X) = \{\mathfrak{p} \mid \Gamma_\mathfrak{p}(X) \neq 0\}$$. Stratification thus reduces the study of subcategories in $D(C^*(BG;k))$ to the prime-ideal geometry of $R = H^*(G, k)$0 (Benson et al., 2011).

This paradigm exemplifies the application of tensor-triangular geometry to modular representation theory, converting triangulated-categorical problems to essentially commutative algebraic ones.

2. Nilpotence and Ring-Theoretic Invariants

Nilpotence in group cohomology is quantified via indices measuring the powers for which the nilradical of the mod $R = H^*(G, k)$1 cohomology ring becomes zero. The algebraic nilpotence index $R = H^*(G, k)$2 and the topological counterpart via unstable module suspensions relate through subgroup invariants, such as the Duflot invariant $R = H^*(G, k)$3.

Symonds’ proof of Benson’s Regularity Conjecture provided a sharp general bound: $R = H^*(G, k)$4 where $R = H^*(G, k)$5 is the nilpotence index and $R = H^*(G, k)$6 is the dimension of $R = H^*(G, k)$7 as a manifold. For finite $R = H^*(G, k)$8-groups, further group-theoretic bounds express $R = H^*(G, k)$9 in terms of the action on faithful sets and the structure of maximal abelian subgroups. These relations refine Quillen’s stratification, revealing deeper structure in the nilradical beyond detection by elementary abelian subgroups (Kuhn, 2010).

3. Enriched and Spectral Data: Bockstein Uniqueness

Bare cohomology rings are often insufficient for full classification of group extensions or spaces up to $D(C^*(BG; k))$0-completion. Incorporation of the Bockstein spectral sequence—a structure recording all higher Bockstein operations—produces a finer invariant within the category $D(C^*(BG; k))$1 of unstable algebras with BSS data.

For certain $D(C^*(BG; k))$2-groups, notably the Leary groups $D(C^*(BG; k))$3, Díaz–Ruiz–Viruel prove that this enriched cohomological datum, comprising both $D(C^*(BG; k))$4 and the BSS, is a complete invariant for p-completed homotopy types. In concrete terms, the differentials in the spectral sequence encode extension classes that distinguish nontrivial central extensions from split versions, even when the underlying cohomology rings coincide (Díaz et al., 2010).

4. Group Cohomology Classification in Algebraic and Kac–Moody Settings

Advanced classification theorems extend to infinite-dimensional and infinite-type groups. For rank 3 Kac–Moody groups, rational and mod $D(C^*(BG; k))$5 cohomology of classifying spaces decompose as sums of Weyl group invariants and their quotients, with precise presentations depending on the parabolic subgroup structure. Nontrivial $D(C^*(BG; k))$6-torsion appears in the integral cohomology for all primes, and the rational ring structure, except for an exceptional case, is always an exterior extension of the polynomial invariants $D(C^*(BG; k))$7 by odd-degree generators (Yangyang et al., 10 Feb 2025).

For unipotent algebraic groups over fields of positive characteristic, the relationship between the cohomology of Frobenius kernels and the rational cohomology is mediated by spectral sequences and universal classes. In the case of the Heisenberg group $D(C^*(BG; k))$8, explicit algebra models—with generators corresponding to roots and central extensions—are constructed and shown to surject onto the full cohomology up to torsion. However, these "universal" classes vanish in the inverse limit to rational cohomology, emphasizing the unique role of Frobenius kernels in positive characteristic (Friedlander, 2017).

5. Continuous Cohomology and Ext-Classifications

For locally profinite groups, continuous group cohomology is classified directly in terms of Ext-groups in the category of smooth representations: $D(C^*(BG; k))$9 This reduction, proven for all smooth representations of a locally profinite group, allows analytic and representation-theoretic techniques to dominate classification in these contexts and enables explicit computation and vanishing results for groups such as $\operatorname{Spec}^h R$0 and their subgroups. The finite-dimensionality of cohomology for admissible Banach representations and the vanishing theorems for supersingular representations highlight this principle in action (Fust, 2021).

6. Limitations and Extensions of Group Cohomology-Based Classifications

Group cohomology $\operatorname{Spec}^h R$1 serves as a first-order approximation for classifying symmetry-protected topological (SPT) phases and related invariants. However, this classification is incomplete: cobordism-theoretic refinements, specifically the Pontryagin duals of torsion in oriented bordism groups over $\operatorname{Spec}^h R$2, capture additional phases—including those invisible to group cohomology—through the technical machinery of the Atiyah–Hirzebruch spectral sequence and explicit computation of bordism invariants. This refined classification governs not only SPT phases but also ’t Hooft anomalies and the structure of gauge/topological actions, especially in higher-dimensional and time-reversal-invariant settings (Kapustin, 2014).

7. Exemplifications and Applications

Concrete group cohomology classifications play decisive roles in computational topology, representation theory, and homotopy theory.

• For cyclic and elementary abelian groups, the classification of localizing subcategories corresponds directly to subsets of affine spaces over $\operatorname{Spec}^h R$3, with canonical computations mediated by Koszul functors and specific algebraic models (Benson et al., 2011).
• For rank 3 Kac–Moody groups, the decomposition by parabolic subgroups determines both the number of summands and their specific algebraic structures, with implications for constructing models of non-finite type group classifying spaces (Yangyang et al., 10 Feb 2025).
• For unipotent and Frobenius kernel settings, concrete algebra models with explicit generators, relations, and supports—such as those for the Heisenberg group—yield algorithms and explicit calculations of group cohomology in positive characteristic (Friedlander, 2017).
• In the context of loop groups and finite Chevalley groups, the cohomology of classifying spaces realizes deep connections between algebraic group cohomology and topological loop space constructions, often producing isomorphisms at the level of A–algebra structures (Kameko, 2014).

A plausible implication is that future developments in group cohomology classification will further integrate categorical, spectral, and geometric frameworks, particularly as new invariants and correspondences (such as cobordism-theoretic or equivariant refinements) are systematically analyzed and classified.