On the continuity of derivations over locally regular Banach algebras

This paper establishes the continuity of derivations over a class of Banach algebras containing dense $C^*$-like subalgebras, specifically demonstrating that every derivation on the $L^p$-crossed product $F^p(G,X,\alpha)$ is continuous when the group $G$ is infinite, finitely generated, has polynomial growth, and acts freely on the compact Hausdorff space $X$.

The Gist

Here is an explanation of the paper, translated from the dense language of advanced mathematics into a story about building, rules, and stability.

The Big Picture: The "Automatic Safety" Problem

Imagine you are an architect designing a massive, complex building (this is a Banach algebra). Inside this building, there are workers called derivations.

In the world of math, a "derivation" is a rule that tells you how to break things down or calculate changes. Think of it like a quality inspector who checks how a structure reacts when you push on it.

• The Rule: If you push on two connected beams ($a$ and $b$), the inspector calculates the stress on the whole ($ab$) by adding the stress on beam $a$ plus the stress on beam $b$. This is the "Leibniz rule."

The Problem:
Sometimes, these inspectors (derivations) are crazy. They might look at a tiny, harmless push and scream that the building is about to collapse. In math terms, they are discontinuous. A tiny change in input causes a massive, infinite change in output. This is bad for stability.

For a long time, mathematicians knew that if the building was a perfect, rigid C-algebra* (like a standard, well-ordered skyscraper), these inspectors were always calm and predictable. They were automatically continuous.

But what about other, weirder buildings? What about Lp-crossed products? These are complex structures built from groups (like symmetries) and spaces. For these, we didn't know if the inspectors would stay calm or go crazy.

The Goal of this Paper:
Felipe Flores wants to prove that for a specific, large class of these "weird" buildings, the inspectors are automatically calm. You don't need to check them individually; the very structure of the building forces them to be stable.

---

The Secret Ingredient: The "C*-like" Skeleton

How does Flores prove this? He uses a clever trick involving a "skeleton" inside the building.

Imagine your weird building (the Banach algebra $B$) is made of a strange, flexible material. However, deep inside it, there is a dense, rigid steel skeleton (a subalgebra $A$).

• This skeleton isn't just any steel; it's "C*-like." It behaves very much like the perfect skyscrapers we already know are safe.
• Crucially, this skeleton is dense. This means if you pick any point in the weird building, you can find a point in the steel skeleton right next to it. The skeleton is everywhere, even if you can't see it all at once.

The "Locally Regular" Condition:
Flores introduces a new concept called a "locally regular inclusion."
Think of this as a guarantee that the steel skeleton is "well-behaved" in every local neighborhood. If you take any single piece of the skeleton and look at the rules governing just that piece, it acts like a perfect, regular function algebra.

The Analogy:
Imagine a chaotic, messy room (the Banach algebra). But, if you look at any single corner of the room, you find a perfectly organized, clean desk (the regular subalgebra). Flores proves that if every corner has a clean desk, and those desks are made of the right kind of steel, then the entire messy room must actually be stable.

---

The Main Results: Two New Theorems

Flores proves two main things using this "skeleton" idea:

1. The "No Tiny Holes" Rule (Theorem 1.1)
If your building has a steel skeleton that is "locally regular," and if the skeleton is so solid that it cannot be broken into tiny, finite pieces (no "finite codimension" ideals), then every inspector inside the building is automatically calm.

• Translation: If the underlying structure is robust and infinite in a specific way, chaos is impossible.

2. The "Bridge" Rule (Theorem 1.2)
This is even more powerful. Imagine you have two different buildings, $B_1$ and $B_2$. They both share the same steel skeleton. If you have an inspector moving from Building 1 to Building 2, and the skeleton is "locally regular," that inspector is automatically calm.

• Translation: You can move between different types of complex algebras (like $L^1$ and $L^2$ versions) without the rules breaking down, as long as they share that same well-behaved core.

---

The Real-World Application: Lp-Crossed Products

Why do we care? Flores applies this to Lp-crossed products.

• These are mathematical objects used to study groups (like symmetries of shapes) acting on spaces (like a city map).
• They are important in physics and geometry, but they are notoriously hard to analyze because they aren't "semisimple" (they have hidden complexities).

The Result:
Flores shows that if you take a group $G$ that is "nice" (it has polynomial growth, meaning it doesn't expand too wildly, like a grid or a tree) and let it act on a space $X$, the resulting algebra is stable.

• The Conclusion: Every derivation (inspector) on these specific algebras is continuous. They won't go crazy.

He also covers Symmetrized Lp-crossed products. These are special versions that have a "mirror" property (an involution), making them even more like the perfect C*-algebras. He proves that even when moving between these different versions (e.g., from $L^1$ to $L^2$), the inspectors remain calm.

---

Summary: Why This Matters

Before this paper, mathematicians had to use different, complicated tools to prove stability for different types of groups. It was like having a different key for every door.

Flores built a Master Key.

1. He realized that if you can find a "C*-like" skeleton inside a messy algebra...
2. ...and that skeleton is "locally regular" (well-behaved everywhere)...
3. ...then the whole algebra is automatically stable.

This unifies many previous results and solves open problems for a huge class of algebras that were previously too messy to handle. It tells us that even in the most complex mathematical structures, if the core is strong and regular, the whole system behaves with automatic order.

Technical Summary

Here is a detailed technical summary of the paper "On the continuity of derivations over locally regular Banach algebras" by Felipe I. Flores.

1. Problem Statement

The central problem addressed is the automatic continuity of derivations on Banach algebras.

• Context: A derivation $D: B \to X$ (where $B$ is a Banach algebra and $X$ is a Banach $B$-bimodule) is a linear map satisfying the Leibniz rule: $D(ab) = D(a)b + aD(b)$.
• The Question: Under what conditions is every such derivation automatically continuous?
• Current State of Knowledge:
  • Ringrose [29] proved that every derivation on a $C^*$-algebra is automatically continuous.
  • Johnson and Sinclair [24] proved continuity for derivations $D: B \to B$ if $B$ is semisimple.
  • However, many important algebras, specifically $L^p$-crossed products ($1 < p < \infty, p \neq 2$), are not known to be semisimple, nor are they$C^*$-algebras. Consequently, existing theorems do not apply to them.
• Goal: To establish automatic continuity for derivations on a broader class of Banach algebras that are not necessarily $C^*$-algebras or semisimple, specifically targeting $L^p$-crossed products.

2. Methodology

The author extends the approach of Longo [25] and Runde [31] by introducing a new structural concept that relaxes the requirements of previous theorems.

Key Concept: Locally Regular Inclusions

The core innovation is the definition of a locally regular inclusion $A \subset B$.

• Setup: $A$ is a dense subalgebra of a Banach algebra $B$. $A$ is an $A^*$-algebra (a Banach $*$-algebra admitting a $C^*$-norm).
• Condition: For every self-adjoint element $b \in A$, the closed subalgebra generated by $b$, denoted $A(b)$, must be a regular Banach function algebra.
• Motivation: Previous results (like Longo's) required a very strong, everywhere-defined functional calculus. Flores demonstrates that a densely defined smooth functional calculus (via the regularity of $A(b)$) is sufficient to carry over the arguments for automatic continuity.
• Mechanism: The proof strategy relies on analyzing the continuity ideal $I(D) = \{b \in B \mid \text{maps } a \mapsto D(ba) \text{ and } a \mapsto D(ab) \text{ are continuous}\}$. The goal is to show that $I(D)$ contains the unit (or is dense enough) to force continuity.

Technical Tools

• Spectral Theory: Utilizing the spectrum of elements in the quotient algebras $A/K$ and $C^*(A)/J$.
• Contradiction via Infinite Dimensions: The proofs often assume the quotient $A/K$ is infinite-dimensional, constructing a sequence of elements with disjoint supports (using the regularity of the function algebra) to derive a contradiction regarding the boundedness of the derivation.
• Interpolation: For $L^p$-crossed products, the paper utilizes complex interpolation (Riesz-Thorin) to relate $L^p$ and $L^q$ norms and establish the existence of continuous involutions.

3. Key Contributions and Main Theorems

Theorem 1.1 (Theorem 2.9)

Statement: Let $A \subset B$ be a locally regular inclusion. Let $D: B \to X$ be a derivation into a Banach $B$-bimodule. If $A$ is unital and the universal $C^*$-algebra $C^*(A)$ has no proper closed two-sided ideals of finite codimension, then $D$ is continuous.

• Significance: This generalizes previous results by removing the requirement that $B$ itself be a $C^*$-algebra or semisimple. It only requires the dense subalgebra $A$ to have specific regularity properties and the $C^*$-envelope to lack finite-codimensional ideals.

Theorem 1.2 (Theorem 2.12)

Statement: Let $A$ be an $A^*$-algebra and $B_1, B_2$ be Banach algebras with continuous injective homomorphisms $A \to B_i$ and $B_i \to C^*(A)$ making the diagrams commute. If $A \to B_1$ is a locally regular inclusion, then every derivation $D: B_1 \to B_2$ is continuous.

• Significance: This applies even if $B_1$ is not unital. It is particularly powerful for mappings between different completions of the same algebra (e.g., from an $L^1$-type algebra to a $C^*$-algebra or another $L^p$-algebra).

4. Applications to $L^p$-Crossed Products

The paper applies these theorems to $L^p$-crossed products and symmetrized $L^p$-crossed products, denoted $F^p(G, X, \alpha)$ and $F^p_*(G, X, \alpha)$.

• Construction: These algebras are completions of the convolution algebra $L^1_\alpha(G, C_0(X))$ under specific norms involving covariant representations on $L^p$ spaces.
• The Subalgebra $E_w$: The author utilizes a weighted subalgebra $E_w$ (functions with polynomial growth weights).
  • Lemma 3.3: If $G$ is a countable locally finite group or a compactly generated group of polynomial growth, there exists a weight $w$ such that $E_w$ is a locally regular inclusion in the relevant completions.
• Corollary 1.3 (Theorem 3.5):
  • Conditions: $G$ is countably infinite, locally finite or finitely generated with polynomial growth. The action $\alpha$ on a compact Hausdorff space $X$ is topologically free on closed invariant subsets.
  • Result: Every derivation $D: F^p(G, X, \alpha) \to X$ (where $X$ is a bimodule) is continuous.
• Corollary 1.4 (Theorem 3.4):
  • Conditions: $G$ is as above, acting on a locally compact Hausdorff space $X$. $p, q \in [1, 2]$.
  • Result: Every derivation $D: F^p_*(G, X, \alpha) \to F^q_*(G, X, \alpha)$ is continuous.
  • Note: This covers cases like $L^1_\alpha(G, C_0(X)) \to F^q_*(G, X, \alpha)$ and $F^p_*(G, X, \alpha) \to F^p_*(G, X, \alpha)$, which were previously inaccessible due to the lack of semisimplicity.

5. Significance and Impact

1. Bridging the Gap: The paper successfully extends automatic continuity results from the realm of $C^*$-algebras and semisimple algebras to the non-semisimple, non-involutive setting of $L^p$-operator algebras.
2. Relaxing Conditions: By introducing "locally regular inclusions," the author bypasses the need for global functional calculus or semisimplicity, relying instead on the density of a well-behaved subalgebra ($A^*$-algebra with regular function subalgebras).
3. Resolution of Open Problems: It provides a definitive answer for the continuity of derivations on $L^p$-crossed products for groups of polynomial growth, a class of algebras where the Johnson-Sinclair theorem fails.
4. Generality: The framework is robust enough to potentially apply to twisted crossed products and algebras of pseudofunctions, suggesting a new standard for analyzing automatic continuity in non-commutative harmonic analysis.

In summary, Flores provides a unified framework that proves the automatic continuity of derivations for a wide class of $L^p$-crossed products by leveraging the regularity of dense $A^*$-subalgebras, thereby solving a significant open problem in the theory of Banach algebras.