Continuity and equivariant dimension

Imagine you are an architect trying to build a house, but instead of wood and bricks, you are building with "mathematical shapes" that live in a strange, non-commutative universe. In this universe, the order in which you do things matters (unlike in our world, where putting on your left shoe then your right is the same as right then left).

This paper by Alexandru Chirvasitu and Benjamin Passer is about measuring the complexity of these mathematical shapes when they are being "twisted" or "rotated" by a group (like a circle spinning or a mirror flipping).

Here is the breakdown of their discovery using everyday analogies.

1. The Core Problem: The "Borsuk-Ulam" Puzzle

Think of the famous Borsuk-Ulam Theorem like this: If you take a globe (a sphere) and paint every point on it, there will always be two opposite points (antipodes) that have the exact same temperature. You can't paint a sphere so that every opposite pair has different temperatures.

In the world of this paper, mathematicians are asking: "How many 'odd' ingredients do we need to build a mathematical structure so that it behaves like a sphere?"

They invented a ruler called the Local-Triviality Dimension.

Low Dimension: The shape is simple. It's easy to "untwist" or flatten out.
High Dimension: The shape is complex. It's very knotted and hard to flatten.
Infinite Dimension: The shape is so twisted it can never be flattened, no matter how hard you try.

2. The Big Surprise: Free vs. Finite

For a long time, mathematicians thought there was a perfect link between two things:

Freeness: The shape is "free" (it doesn't get stuck or tangled when you rotate it).
Finite Dimension: The complexity ruler shows a specific, finite number.

The Paper's Discovery: They found that a shape can be "free" (untangled) but still have an "infinite" complexity ruler.

The Analogy: Imagine a rubber band.

If you stretch it, it's "free" (it moves without sticking).
Usually, a simple rubber band has a low "complexity score."
But Chirvasitu and Passer found a rubber band that is perfectly free to move, yet it is twisted in such a weird, infinite way that its complexity score is infinity. This breaks the old rule that "free means simple."

3. The "Continuous Field" Mystery

The authors also studied what happens when you have a family of these shapes that change smoothly, like a movie reel where each frame is a slightly different shape.

The Expectation: If you have a smooth movie of shapes, you'd expect the "complexity score" to change smoothly too. If Frame 1 has a score of 3, and Frame 2 has a score of 4, Frame 1.5 should be 3.5.
The Reality: The paper shows that the complexity score can jump.
- Imagine a movie where the shape is a simple circle (Score: 0) for most of the time.
- But right at the very end, it suddenly snaps into a complex knot (Score: 3).
- The score didn't drift up; it jumped.

They call this Upper Semicontinuity. It means the score can drop suddenly, but it can't jump up suddenly unless you are looking at a very specific type of twist. In many cases, the "average" complexity of the whole movie is actually lower than the complexity of the individual frames.

4. The Special Cases: Noncommutative Tori and Spheres

The authors tested their theories on two specific types of mathematical shapes:

Noncommutative Tori: Think of a donut where the surface is made of "quantum pixels" that don't follow normal rules.
Noncommutative Spheres: Think of a ball where the coordinates are fuzzy and order-dependent.

They looked at what happens when you change a "knob" (a parameter called $\theta$ ) that controls how fuzzy the shape is.

Rational Numbers: When the knob is set to a simple fraction (like 1/2 or 3/4), the shape behaves like a bundle of smaller, simpler shapes (like a bundle of sticks).
Irrational Numbers: When the knob is set to a weird number (like $\pi$ ), the shape becomes a single, solid, infinitely complex knot.

The Result: As you turn the knob from a simple fraction to an irrational number, the complexity score doesn't drift. It stays low for the fractions, and then SNAPS to infinity the moment you hit an irrational number.

Summary of the "Takeaway"

Freeness $\neq$ Simplicity: Just because a mathematical shape is "free" (doesn't get stuck) doesn't mean it's simple. It can be infinitely complex.
Smooth Changes aren't Smooth: If you have a family of shapes changing smoothly, their "complexity score" might jump around wildly. You can't always predict the complexity of the whole family just by looking at the individual pieces.
The "Jump": In the world of quantum shapes, small changes in the rules can cause the complexity to explode from zero to infinity instantly.

In a nutshell: The authors mapped the "terrain" of these strange mathematical shapes and found that the landscape is full of cliffs and jumps, not just gentle hills. What we thought was a smooth path is actually full of sudden drops and infinite peaks.

Here is a detailed technical summary of the paper "Continuity and equivariant dimension" by Alexandru Chirvasitu and Benjamin Passer.

1. Problem Statement

The paper investigates the behavior of local-triviality dimensions (LT dimensions) for actions on $C^*$ -algebras, specifically within the context of noncommutative Borsuk-Ulam theory.

Context: The classical Borsuk-Ulam Theorem states that there is no continuous odd function from $S^n$ to $S^m$ if $m < n$ . In the noncommutative setting, this translates to conditions on the "freeness" of group actions on $C^*$ -algebras and the existence of equivariant morphisms.
The Gap: While it is known that the finiteness of local-triviality dimensions implies the freeness of an action, the converse is not always true. Furthermore, the behavior of these dimensions under deformations (specifically continuous fields of $C^*$ -algebras) is poorly understood.
Key Questions:
1. Do free actions always possess finite weak local-triviality dimensions?
2. How do LT dimensions behave across the fibers of a continuous field of $C^*$ -algebras? Are they continuous, or do they exhibit semicontinuity?
3. Can the dimensions of a global field exceed the supremum of the dimensions of its individual fibers?

2. Methodology

The authors employ a blend of operator algebra theory, noncommutative topology, and bundle theory.

Definitions and Frameworks:
- They utilize three variants of local-triviality dimensions: Weak ( $dim_{WLT}$ ), Plain ( $dim_{LT}$ ), and Strong ( $dim_{SLT}$ ). These are defined via the existence of equivariant $*$ -homomorphisms from specific tensor products involving the group algebra and the interval $(0,1]$ .
- They focus on actions of compact abelian groups (specifically $S^1$ and its finite cyclic subgroups $\mathbb{Z}/k$ ) on unital $C^*$ -algebras.
Continuous Fields: The study centers on $C(X)$ -algebras, where $X$ is a compact Hausdorff space. The authors analyze how dimensions vary as one moves through the fibers $A_x$ of a field $A$ .
Specific Objects:
- $\theta$ -Spheres: Noncommutative spheres $C(S^{2n-1}_\theta)$ defined by Natsume and Olsen, where generators satisfy scaled commutation relations.
- Noncommutative Tori: $C^*$ -algebras $A^n_\theta$ generated by unitaries with commutation relations determined by an antisymmetric matrix $\theta$ .
- Matrix Bundles: The authors utilize the realization of rational noncommutative tori as sections of vector bundles (specifically endomorphism bundles of rank- $q$ vector bundles over the torus).
Topological Tools:
- Index Theory: They define indices for vector and matrix bundles (related to the numerical $G$ -index) to bound the dimensions.
- Homotopy and Flag Manifolds: In proving lower bounds for strong dimensions, they construct maps to flag manifolds and analyze fundamental groups to derive contradictions.
- Partition of Unity: Used to patch local sections of bundles into global sections to establish upper bounds on dimensions.

3. Key Contributions and Results

A. Freeness vs. Finite Dimensions

The authors demonstrate that freeness of an action does not guarantee finite weak local-triviality dimension.

Counterexample: They construct a free action of $(\mathbb{Z}/q)^2$ on the matrix algebra $M_q$ (Example 3.22). While the action is free (satisfying the strong grading condition), the weak local-triviality dimension is infinite. This refutes the idea that freeness and finite LT dimensions are equivalent in the general noncommutative setting.

B. Semicontinuity and Discontinuity in Fields

The paper establishes that LT dimensions are generally upper semicontinuous but not continuous with respect to deformation parameters.

Upper Semicontinuity: For actions of groups like $\mathbb{Z}/2^m$ , the function $x \mapsto dim_{WLT}(A_x)$ is upper semicontinuous (Theorem 3.17). This means the dimension can jump up at a limit point but cannot jump down.
Discontinuity Examples:
- $\theta$ -Spheres: For the antipodal action on $\theta$ -deformed spheres, $dim_{WLT}$ is 3 at the commutative limit ( $\theta=0$ ) but drops to 1 for non-integral $\theta$ .
- Berezin Quantization: A field where fibers are matrix algebras $M_n$ (dimension 0) converging to $C(S^2)$ (dimension $\infty$ ) under a specific $\mathbb{Z}/2$ action.
- Noncommutative Tori: For ergodic actions of $T^2$ on $A_\theta$ , dimensions are 0 at integral $\theta$ but infinite for non-integral $\theta$ .

C. Global vs. Fiber Dimensions

The authors prove that the global dimension of a field can be strictly greater than the supremum of the dimensions of its fibers.

Example 3.7: A twisted action on $C(\mathbb{C}P^1) \otimes M_2$ . The fiber-wise dimensions are 0, but the global weak LT dimension is 1. This is due to topological obstructions (related to the non-triviality of the universal subbundle on $\mathbb{C}P^1$ ) preventing the patching of local trivializations.

D. Rational Tori and Strong Dimensions

A significant portion of the paper focuses on rational noncommutative tori ( $\theta = p/q$ ).

Bundle Realization: They utilize the fact that $A_\theta \cong \Gamma(E \otimes E^*)$ for a rank- $q$ vector bundle $E$ .
Strong Dimension Lower Bound: They prove that for rational $\theta$ with odd denominators, the strong local-triviality dimension of the $\theta$ -sphere is exactly $k$ (Theorem 4.3).
Infinite Strong Dimension: For rational $\theta = p/q$ , the strong local-triviality dimension of the $\mathbb{Z}/q$ -action on the sphere is infinite (Proposition 4.7). The proof involves showing that a finite dimension would imply the existence of a section of a specific bundle over the torus that contradicts homotopy properties of the flag manifold $U(q)/U(1)^q$ .

4. Significance

Refinement of Noncommutative Borsuk-Ulam Theory: The paper clarifies the relationship between "freeness" and "dimension." It shows that while finite dimension implies freeness, the converse fails, necessitating a more nuanced understanding of noncommutative topology.
Pathology of Continuous Fields: It highlights that invariants in noncommutative geometry (like LT dimensions) do not behave as smoothly as their commutative counterparts. The failure of continuity and the strict inequality between global and fiber dimensions reveal deep topological obstructions inherent in noncommutative bundles.
Computational Framework: By linking LT dimensions to indices of matrix bundles and quotient spreads, the authors provide concrete computational tools for determining these invariants in specific cases (like rational tori).
Counterexamples: The construction of free actions with infinite dimensions and continuous fields with discontinuous dimensions provides essential counterexamples that define the boundaries of current theorems in the field.

In summary, Chirvasitu and Passer provide a rigorous analysis of the stability and continuity of equivariant dimensions, demonstrating that noncommutative deformations can drastically alter topological invariants and that the "freeness" of an action is a weaker condition than the finiteness of its associated dimension.