On the homology of partial group representations

Basic Information

• Paper ID: 2404.14650
• Title: On the homology of partial group representations
• Author: Emmanuel Jerez (Departamento de Matemática, Universidade de São Paulo, Brazil)
• Classification: math.AT (Algebraic Topology), math.GR (Group Theory), math.KT (K-Theory and Homological Algebra)
• Publication Date: April 2024 (arXiv preprint, revised October 2025)
• Paper Link: https://arxiv.org/abs/2404.14650

Abstract

This paper investigates how partial group (co)homology of a group $G$ can be described through classical group (co)homology. The author introduces the concept of universal globalization $\Lambda(M)$ for partial group representations $M$. The main result establishes a natural isomorphism between partial group homology $H^{\text{par}}_{\bullet}(G, M)$ and classical group homology $H_{\bullet}(G, \Lambda(M))$. In the cohomological framework, the author obtains a spectral sequence converging to partial group cohomology. When $G$ is countable, the spectral sequence degenerates, yielding a natural isomorphism $H^{\bullet}_{\text{par}}(G, M) \cong H^{\bullet}(G, \operatorname{Hom}_{K_{\text{par}} G}(\Lambda(K_{\text{par}}G), M))$.

Research Background and Motivation

Problem Context

1. Importance of partial group actions: Partial group actions were introduced by Exel in C*-algebra theory to encode symmetries defined only on subsets of a space. These structures naturally arise when studying local behavior of families of symmetries.
2. Absence of homological theory: While the theory of partial group actions has matured, there is a lack of appropriate (co)homology theories to study the structure of partial group actions. Existing preliminary approaches include cohomology based on semigroups and cohomology based on partial representations.
3. Computational challenges: Developing methods to compute partial group (co)homology is crucial for understanding the structure of partial group actions and partial representations, and is also important for computing Hochschild (co)homology of partial twisted group algebras.

Core Problem

Main question: Can one utilize classical group (co)homology of a group $G$ to compute partial group (co)homology of $G$ with coefficients in partial representations? If so, what is the relationship between these two homological theories?

Research Significance

1. Theoretical bridge: Establishes connections between classical and partial homological invariants, enabling the use of mature group cohomology tools to analyze partial structures.
2. Computational advantage: The standard complex for partial group (co)homology is smaller than that for classical group (co)homology, providing computational benefits.
3. Applied value: Results encompass group algebras within the Hopf algebra class, providing an application instance for 3, Proposition 5.2.

Core Contributions

1. Introduction of universal globalization: Constructs universal globalization $\Lambda(M)$ for arbitrary partial group representations $M$, establishing a functor from partial representations to global representations.
2. Establishment of homological isomorphism: Proves natural isomorphism between partial group homology $H^{\text{par}}_{\bullet}(G,M)$ and classical group homology $H_{\bullet}(G,\Lambda(M))$.
3. Construction of cohomological spectral sequence: Obtains a spectral sequence converging to partial group cohomology in the cohomological framework; the spectral sequence degenerates when $G$ is countable.
4. Resolution of open conjecture: Provides an affirmative answer to 2, Conjecture D for countable groups.
5. Development of partial tensor product theory: Constructs partial tensor products for partial group actions and proves they satisfy common properties of tensor products.

Detailed Methodology

Problem Formulation

Study the homological properties of partial group representations, specifically establishing relationships between partial group (co)homology and classical group (co)homology.

Input: A group $G$ and a partial group representation $M$ of $G$ Output: Partial group (co)homology groups and their relationship with classical group (co)homology

Core Constructions

1. Partial group algebras and partial representations

• Exel semigroup: $S(G)$ is the inverse semigroup defined by generators $[t], t \in G$ and relations
• Partial group algebra: $K_{\text{par}}G$ is the semigroup $K$-algebra generated by $S(G)$
• Partial representation: A map $\pi: G \to \text{End}_K(M)$ satisfying specific compatibility conditions

2. Partial tensor product construction

For a right $G_{\text{par}}$-module $X$ and left $G_{\text{par}}$-module $Y$, define the partial tensor product: $X \otimes_{G_{\text{par}}} Y := \frac{X \otimes_K Y}{K_{\beta,\alpha}}$

where $K_{\beta,\alpha}$ is the submodule generated by elements $\{(x)\beta_g \otimes y - x \otimes \alpha_g(y)\}$.

3. Universal globalization

Theorem: For a partial group action $\alpha: G \curvearrowright M$, the triple $(KG \otimes_{G_{\text{par}}} M, \Theta, \iota)$ constitutes a universal global action, where:

• $\Theta: G \curvearrowright KG \otimes_{G_{\text{par}}} M$ is the associated global action
• $\iota: M \to KG \otimes_{G_{\text{par}}} M, m \mapsto 1 \otimes_{G_{\text{par}}} m$

Technical Innovations

1. Exploitation of universal properties

Proves existence and uniqueness of the globalization functor $\Lambda := KG \otimes_{G_{\text{par}}} -$ through universal properties, avoiding the complexity of direct construction.

2. Introduction of partial tensor product

The partial tensor product not only provides technical tools but also satisfies fundamental properties of tensor products such as associativity and universal properties, establishing a solid foundation for the theory.

3. Case-by-case analysis

Different technical approaches are employed for different types of groups (finite, countable, uncountable):

• Countable groups: Constructive proof of projectivity
• Uncountable groups: Proof of non-projectivity
• Finite groups: Exploitation of finite presentation properties

Main Results

Homological Isomorphism Theorem

Theorem 3.7: Let $\pi: G \to \text{End}_K(M)$ be a partial group representation. Then $H^{\text{par}}_{\bullet}(G,M) \cong H_{\bullet}(G,\Lambda(M))$

Proof strategy:

1. Utilize the fact that $KG \otimes_{G_{\text{par}}} K_{\text{par}}G$ is free as a left $KG$-module
2. Construct a flat resolution $P_{\bullet} \otimes_{KG} (KG \otimes_{G_{\text{par}}} K_{\text{par}}G) \to B$
3. Apply homological computation to obtain the isomorphism

Cohomological Spectral Sequence

Theorem 3.22: For a right $K_{\text{par}}G$-module $M$, there exists a cohomological spectral sequence $E_2^{p,q} = H^p(G, \text{Ext}^q_{K_{\text{par}}G}(\Lambda(K_{\text{par}}G), M)) \Rightarrow H^{p+q}_{\text{par}}(G,M)$

When $G$ is finite or countable, the spectral sequence degenerates to yield: $H^n_{\text{par}}(G,M) \cong H^n(G, \text{Hom}_{K_{\text{par}}G}(\Lambda(K_{\text{par}}G), M))$

Projectivity Results

Proposition 3.17: If $G$ is uncountably infinite, then $KG \otimes_{G_{\text{par}}} K_{\text{par}}G$ is not projective as a right $K_{\text{par}}G$-module.

Proposition 3.19: If $G$ is countably infinite, then $KG \otimes B$ is projective as a right $K_{\text{par}}G$-module.

Applications and Corollaries

Lyndon-Hochschild-Serre Spectral Sequence

Corollary 3.8: Let $N$ be a normal subgroup of $G$. Then there exists a spectral sequence $H_p(G/N, H_q(N, \Lambda(M))) \Rightarrow H^{\text{par}}_{p+q}(G,M)$

Shapiro Lemma

Corollary 3.9: Let $S$ be a subgroup of $G$, and $\pi: S \to \text{End}_K(M)$ be a partial representation. Then $H^{\text{par}}_{\bullet}(S,M) \cong H_{\bullet}(G, KG \otimes_{S_{\text{par}}} M)$

Cohomological Dimension

Corollary 3.25:

• If $G$ is finite or countable, then $\text{cd}^{\text{par}}_K(G) = \text{cd}_K(G)$
• If $G$ is uncountably infinite, then $\text{cd}^{\text{par}}_K(G) \leq \text{cd}_K(G) + \text{pd}_{K_{\text{par}}G}(KG \otimes_{G_{\text{par}}} K_{\text{par}}G)$

Technical Details

Key Lemmas

Lemma 3.5: $KG \otimes_{G_{\text{par}}} K_{\text{par}}G$ is isomorphic to $KG \otimes B$ as left $KG$-modules; in particular, the former is free.

Lemma 3.16: Let $G$ be infinite and $z \in K_{\text{par}}G$. If there exists an infinite set $S \subseteq G$ such that $ze_g = 0$ for all $g \in S$, then $z = 0$.

Construction Techniques

In the proof of projectivity for countable groups, the author employs a recursive construction method, inductively defining a sequence $\{x_n\}$ satisfying specific compatibility conditions. This construction method exhibits certain technical difficulty and originality.

Related Work

1. Exel's theory of partial actions: Provides foundational framework for partial group actions
2. Alvares et al.'s cohomology theory: Cohomology theory constructed based on partial representations
3. Dokuchaev and Khrypchenko's partial cohomology: Semigroup-based cohomological approach
4. Expansion theory for Hopf algebras: Results in this paper relate to the expansion functor in 3

Conclusions and Discussion

Main Conclusions

1. Establishes precise relationships between partial group (co)homology and classical group (co)homology
2. Proves existence and functoriality of universal globalization
3. Completely characterizes degeneration conditions for spectral sequences

Limitations

1. Complexity for uncountable groups: For uncountable groups, spectral sequences may not degenerate, making relationships more complex
2. Computational complexity: While theoretically establishing connections, actual computation may still face difficulties
3. Restrictions on coefficient rings: Main results are established over commutative unital rings; generalization to non-commutative settings requires further research

Future Directions

1. Generalization to general Hopf algebras: Extend techniques to partial Hopf algebra representations
2. Non-commutative base rings: Study analogous results over non-commutative base rings
3. Geometric applications: Explore applications in algebraic geometry and topology

In-Depth Evaluation

Strengths

1. Theoretical depth: Establishes profound connections between partial group theory and classical group theory
2. Technical innovation: The constructions of partial tensor products and universal globalization exhibit originality
3. Completeness: Addresses both homological and cohomological cases, providing a complete theoretical framework
4. Applied value: Resolves several open problems in the field

Weaknesses

1. Technical complexity: Certain proofs (particularly projectivity for countable groups) are highly technical with limited readability
2. Insufficient examples: Lacks concrete computational examples illustrating practical applications of the theory
3. Lack of geometric intuition: Missing geometric or topological intuitive explanations

Impact

1. Theoretical contribution: Provides important homological tools for partial group theory
2. Computational tools: Offers new computational methods for researchers in related fields
3. Open problems: Resolves important conjectures in the field, advancing theoretical development

Applicable Scenarios

1. C-algebra theory*: Computing K-theory and cohomology of partial crossed products
2. Non-commutative geometry: Studying partial symmetries of non-commutative spaces
3. Representation theory: Analyzing homological properties of partial group representations

References

The paper cites 30 important references, primarily including:

• 10,11 Foundational theory of Exel's partial actions
• 1,2 Prior work on partial group cohomology
• 13,14,15 Standard references in homological algebra
• 3 Expansion theory for partial representations of Hopf algebras

Overall Assessment: This is a high-quality theoretical mathematics paper establishing important connections between partial group theory and classical group theory. While technically demanding, it makes significant theoretical contributions and provides important tools for the field's development. The paper's primary value lies in its theoretical depth and completeness, establishing a solid foundation for subsequent research.