The coordinate change formula for the Liouville quantum gravity metric holds for all conformal maps simultaneously

This paper proves that the coordinate change formula for the Liouville quantum gravity distance function holds almost surely for all conformal maps simultaneously, thereby rigorously establishing the definition of a quantum surface as a random equivalence class of domains equipped with both LQG area and distance measures.

The Gist

Imagine you are looking at a piece of crumpled, shimmering fabric. This fabric represents Liouville Quantum Gravity (LQG). In the world of physics and mathematics, this isn't just a piece of cloth; it's a model for a "random universe" where space itself is constantly jiggling, stretching, and warping due to quantum fluctuations.

The paper you provided is a major breakthrough in understanding how to measure distances on this wobbly fabric. Here is the story of what the author, Charles Devlin VI, actually did, explained without the heavy math.

The Problem: The "Shape-Shifting" Universe

In our normal world, if you have a map of a city and you stretch the paper, the distance between two points changes. But if you know how you stretched the paper (the math behind the stretch), you can calculate the new distance perfectly.

In the quantum world of LQG, the "paper" (the universe) is not just being stretched by a human hand; it is being distorted by a chaotic, random force (the Gaussian Free Field).

• The Old Rule: Mathematicians knew how to calculate distances if you changed the map one specific way at a time.
• The Big Question: What if you wanted to change the map in every possible way at the exact same time? Could you say, "No matter how I twist, turn, or stretch this random universe, the rules for measuring distance still hold true simultaneously for every single version?"

For a long time, this was a "heuristic" (a guess based on intuition). We knew it worked for the area (how much space a region takes up), but nobody could prove it worked for the distance (how far it is to walk from A to B) for all changes at once.

The Solution: The "Universal Translator"

This paper proves that yes, it works. The author constructed a "Universal Translator" for distances.

Here is the analogy:
Imagine you have a magical, bumpy terrain (the LQG surface).

1. The Map Makers: There are infinite artists, each drawing a different map of this terrain. Some draw it stretched, some squished, some rotated.
2. The Ruler: In the quantum world, a standard ruler doesn't work because the ground is shifting. You need a "Quantum Ruler" that adapts to the terrain.
3. The Discovery: Devlin proved that if you use this Quantum Ruler, all the artists' maps are consistent with each other at the same time.

If Artist A says "It's 5 steps from the mountain to the river," and Artist B (who drew a distorted map) says "It's 5 steps," they are both right, even though their maps look totally different. The paper proves that the "5 steps" is a fundamental truth of the universe, not an artifact of how you drew the map.

How Did He Do It? (The "Microscope" Strategy)

Proving this for every possible map at once is incredibly hard because the math gets messy when you try to look at the whole picture. Devlin used a clever strategy called Multiscale Analysis.

Think of it like fixing a giant, tangled knot of yarn:

1. Zoom In (The Microscope): First, he looked at the terrain through a microscope at a tiny, tiny scale. At this level, the wild, random bumps look almost flat and predictable. He proved that for these tiny dots, the distance rules hold up perfectly, no matter how the map is twisted.
2. The "Good Annulus" (The Safe Zone): He realized that even though the whole terrain is chaotic, there are many small rings (like tree rings) where the terrain is "well-behaved." He proved that if you walk from point A to point B, you will inevitably pass through enough of these "safe rings" to make the journey predictable.
3. Stringing It Together (The Ladder): He didn't try to prove the rule for the whole journey at once. Instead, he showed that if the rule works for the tiny rings, and you have enough of them, you can "string them together" like rungs on a ladder to prove the rule works for the whole journey.
4. Iterating (The Polish): He did this over and over again. Each time he did it, the "error" in his measurement got smaller and smaller, until the error vanished completely.

Why Does This Matter?

Before this paper, a "Quantum Surface" was a bit of a fuzzy concept. It was like saying, "This is a random shape, and we hope our math works."

Now, thanks to this paper, a Quantum Surface is a rigorous, solid mathematical object.

• It defines a "Class": Just as a circle is the same shape whether it's drawn big or small, a Quantum Surface is a specific "class" of shapes.
• It unifies the view: It allows physicists and mathematicians to treat these random universes as stable objects that can be studied, compared, and used to understand the fundamental nature of space and time, without worrying that their results depend on which "map" they happened to pick.

The Bottom Line

Charles Devlin VI took a chaotic, random, and seemingly impossible-to-measure quantum universe and proved that distance is a universal constant within it. No matter how you twist the universe, the rules of how far apart things are remain consistent, provided you use the right quantum ruler. It turns a "fuzzy guess" into a "hard fact."

Technical Summary

Here is a detailed technical summary of the paper "The coordinate change formula for the Liouville quantum gravity metric holds for all conformal maps simultaneously" by Charles Devlin VI.

1. Problem Statement and Context

Liouville Quantum Gravity (LQG) is a theoretical framework describing a random 2-dimensional Riemannian manifold with a metric tensor formally given by $e^{\gamma h}(dx^2 + dy^2)$, where $h$ is a Gaussian Free Field (GFF) and $\gamma \in (0, 2)$. Since $h$ is a distribution rather than a function, the metric cannot be defined pointwise. Instead, rigorous constructions define LQG via two key observables:

1. The Area Measure ($\mu$): Constructed as a limit of regularized exponentials of the field.
2. The Distance Function (Metric, $D_h$): Constructed as the limit of Liouville First Passage Percolation (LFPP).

A fundamental heuristic in LQG is that a "quantum surface" is an equivalence class of domains under conformal transformations. Specifically, if $\phi: U \to \phi(U)$ is a conformal map, the field transforms as $h_\phi = h \circ \phi^{-1} + Q \log |(\phi^{-1})'|$ (where $Q = \frac{2}{\gamma} + \frac{\gamma}{2}$). The Coordinate Change Formula states that the LQG metric on $U$ defined by $h$ should be equivalent to the LQG metric on $\phi(U)$ defined by $h_\phi$.

The Gap:
While the coordinate change formula for the area measure was proven to hold almost surely for all conformal maps simultaneously (Sheffield-Wang, 2016), the corresponding result for the distance function remained unproven. Previous results (e.g., Gwynne-Miller, Devlin) established the formula for a fixed conformal map or for affine transformations, but the event of validity could depend on the specific map $\phi$. This prevents a rigorous definition of LQG surfaces as random equivalence classes where all parameterizations are equivalent simultaneously.

The Goal:
To prove that there exists a version of the LQG metric such that the coordinate change formula holds almost surely for all conformal maps $\phi$ simultaneously.

2. Methodology

The proof relies on a multiscale analysis of the Liouville First Passage Percolation (LFPP) metric, denoted $D_\epsilon^h$, which approximates the LQG metric $D_h$ as $\epsilon \to 0$. The core strategy involves proving the simultaneous convergence of coordinate-changed LFPP metrics to the same limit.

Key Technical Components:

1. Localized LFPP:
   Standard LFPP uses heat kernel mollification, which is non-local (depends on the field everywhere). To utilize the local independence properties of the GFF (crucial for handling infinite families of maps), the author introduces a localized variant $\hat{D}_\epsilon^h$. This uses a truncated mollifier supported on a small ball, ensuring the metric depends only on the field in a local neighborhood.

2. Small-Scale Comparison (Section 3):
   The author compares the mollified field $h_\phi$ under a conformal map $\phi$ with the original field $h$ near a point $z_0$.

   • Using distortion estimates for conformal maps (Koebe 1/4 theorem, Taylor expansions), the paper shows that for small scales, $\phi$ behaves like an affine map $z \mapsto az+b$.
   • It is proven that the mollified fields and the resulting path lengths are close to those of the affine transformation, uniformly over a class of conformal maps $\Lambda_\tau(V, U)$ (maps with bounded derivative distortion).
   • Proposition 3.1 establishes that on small scales, the metric $a_\epsilon^{-1} \hat{D}_\epsilon^{h_\phi}(\phi(\cdot), \phi(\cdot))$ is within a factor of $(1+\delta)$ of the scaled metric $a_{\epsilon/|a|}^{-1} \hat{D}_{\epsilon/|a|}^h(\cdot, \cdot)$, where $a = \phi'(z_0)$.

3. Multiscale Iteration and Lipschitz Improvement (Section 4):
   The small-scale estimate is insufficient for global convergence. The paper employs a multiscale argument inspired by previous works (Gwynne-Miller, Devlin) but adapted for a uniform family of maps.

   • Initial Lipschitz Condition: Using the local independence of the GFF on disjoint annuli, the author shows that the coordinate-changed metric and the limit metric $D_h$ are globally "almost Lipschitz equivalent" with some constant $C_0$.
   • Iterative Improvement: The core innovation is an iterative scheme (Proposition 4.10). By identifying "good annuli" where the small-scale estimate holds with high probability, the author shows that a positive proportion of any geodesic lies in these regions.
   • Scaling Ratios: The iteration involves scaling ratios of the form $\frac{r a_\epsilon^{-1}}{r^{\xi Q} a_{\epsilon/r}^{-1}}$. Using the regular variation of the normalization constants $a_\epsilon$, the author proves (Lemma 4.20) that one can choose radii such that these ratios are arbitrarily close to 1.
   • Convergence: By iterating this process, the Lipschitz constant is driven arbitrarily close to 1. This implies that the coordinate-changed metrics converge to the same limit $D_h$ simultaneously for all $\phi$.

4. Handling Infinite Families:
   A major technical hurdle is that the set of conformal maps is uncountable. The proof uses:

   • Separability: Approximating the space of conformal maps with a countable dense subset (Lemma 4.5).
   • Union Bounds: Carefully managing probability bounds over the countable set and using Borel-Cantelli arguments along a discrete sequence $\epsilon = 2^{-n}$.

3. Key Contributions and Results

Main Theorem (Theorem 1.1):
For $\xi < \xi_{crit}$ (the subcritical and critical regimes), there exists a version of the LQG metric $(D_U)_{U \subset \mathbb{C}}$ satisfying the Strong Coordinate Change Formula:
For any open set $U$, any whole-plane GFF $h$ (plus continuous function), and almost surely for all conformal maps $\phi: U \to \phi(U)$:
$$D_U^h(z, w) = D_{\phi(U)}^{h_\phi}(\phi(z), \phi(w)) \quad \forall z, w \in U$$
where $h_\phi = h \circ \phi^{-1} + Q \log |(\phi^{-1})'|$.

Specific Contributions:

1. Simultaneity: The result holds for all conformal maps simultaneously, not just a fixed one or a countable set. This resolves the ambiguity in defining LQG surfaces as equivalence classes.
2. Localized LFPP: The introduction and rigorous analysis of the localized LFPP metric $\hat{D}_\epsilon^h$ to leverage GFF independence properties.
3. Uniform Estimates: The derivation of distortion estimates and scaling ratio controls that are uniform over the class of conformal maps with bounded derivative distortion.
4. Iterative Lipschitz Scheme: A refined version of the iterative Lipschitz improvement argument that accounts for the dependence on the conformal map's derivative.

4. Significance

• Rigorous Foundation for Quantum Surfaces: This paper provides the final rigorous justification for the heuristic definition of an LQG surface as a random equivalence class of domains. It confirms that the geometric structure (distance and area) is intrinsic to the surface and independent of the coordinate chart used to describe it.
• Uniqueness of the Metric: Combined with previous uniqueness results (Gwynne-Miller), this solidifies the LQG metric as the unique object satisfying the natural axioms (Length space, Locality, Weyl scaling, and now Strong Coordinate Change).
• Methodological Advancement: The techniques developed for handling simultaneous convergence over infinite families of maps and the use of localized mollifiers are likely to be applicable to other problems in random geometry and stochastic analysis involving conformal invariance.
• Completeness of the Theory: It bridges the gap between the area measure (Sheffield-Wang) and the distance function, completing the rigorous definition of the two primary observables of LQG.

In summary, Devlin VI successfully extends the known coordinate change properties of the LQG area measure to the more complex LQG distance function, proving that the metric structure of Liouville Quantum Gravity is robustly conformally invariant across all possible parameterizations simultaneously.