On the Onsager-Machlup functional of the $Φ^4$-measure

This paper investigates the existence and form of Onsager-Machlup functionals for $\Phi^4_d$ measures in dimensions $d=1,2,3$, demonstrating that they recover the standard action in one and two dimensions (using enhanced distances) while proving degeneracy in three dimensions unless specific joint limits are considered.

The Gist

Imagine you are trying to describe the "most likely shape" of a wiggly, invisible string floating in a room. In physics, this string is a quantum field, and the room is a mathematical space called a "torus" (think of a donut shape).

Physicists have a formula, called the Action ($S$), that tells them how much "effort" or "energy" it takes for the string to take a specific shape. Usually, they say the probability of finding the string in a certain shape is proportional to $e^{-\text{Energy}}$. It's like saying: "The higher the energy, the less likely you are to see that shape."

However, there's a catch. In the world of quantum fields (especially in 2 and 3 dimensions), the string is so wild and jagged that it's not a smooth line at all; it's a chaotic cloud of dust. Because it's so jagged, the standard math tools used to calculate probabilities (like measuring the volume of a ball) break down. You can't just say "the probability of being in this tiny ball" because the ball might contain infinite energy or undefined numbers.

This paper is about finding a new, better way to measure these probabilities for these wild strings. The authors use a tool called the Onsager-Machlup (OM) functional.

The Analogy: The "Tiny Ball" Test

Imagine you want to know which shape of the string is the "most likely."

1. The Standard Way: You draw a tiny circle (a ball) around a specific shape $A$ and another tiny circle around shape $B$. You count how many times the string appears in circle $A$ versus circle $B$.
2. The Problem: In 3D, the string is so jagged that if you draw a standard circle, it might be empty (probability 0) for almost every shape you pick, or the math explodes.
3. The Solution (The OM Functional): Instead of just measuring the size of the circle, we look at the ratio of probabilities. If the ratio of "Probability of A" to "Probability of B" approaches a specific number as the circles get infinitely small, that number tells us the "energy cost" (the Action) of shape A compared to shape B.

What the Authors Found (Dimension by Dimension)

The authors tested this method in three different "worlds" (dimensions): 1D, 2D, and 3D.

1. The Easy World (1 Dimension)

• The Situation: In 1D, the string is just a line. It's a bit wiggly, but not crazy.
• The Result: The authors checked the "tiny ball" ratio, and it worked perfectly. The math matched the standard energy formula exactly.
• The Metaphor: It's like measuring the weight of a smooth pebble. You put it on a scale, and it gives you the right number. No tricks needed.

2. The Tricky World (2 Dimensions)

• The Situation: In 2D, the string is a surface. It's getting very jagged. If you try to measure it with a standard "ball," the jagged edges ruin the measurement.
• The Fix: The authors realized that to measure this properly, you can't just look at the shape itself. You have to look at the shape plus its "roughness features" (mathematically called Wick powers).
• The Metaphor: Imagine trying to measure a crumpled piece of paper. If you just look at the outline, you get confused. But if you also measure how crumpled it is (the "roughness"), you can finally get a consistent measurement.
• The Result: By using these "enhanced" balls that check both the shape and the roughness, they found that the probability ratio did match the standard energy formula. They successfully recovered the "Action."

3. The Impossible World (3 Dimensions)

• The Situation: In 3D, the string is a volume. It is so incredibly wild that it's like trying to measure a cloud of smoke with a ruler.
• The Problem: The authors tried the same "enhanced" trick they used in 2D. They made "super-balls" that checked the shape and all its roughness features.
• The Result: It failed. No matter how they tried, the math broke.
  • If they picked a "smooth" shape (like a flat plane), the probability of the wild string looking like that smooth shape was zero.
  • If they picked a "wild" shape, the probability was also effectively zero or undefined.
  • The Metaphor: It's like trying to find a perfect, smooth sphere inside a tornado. The tornado (the quantum field) is so chaotic that it simply never looks like a smooth sphere, no matter how small you look. The "Action" formula becomes useless because the two things being compared are mutually exclusive (they can't coexist).
• The Conclusion: In 3D, the standard "density" formula doesn't exist in a simple way. The "most likely path" concept breaks down.

The "Hail Mary" Pass (Section 6)

Even though the 3D result was a "failure" (the functional was degenerate), the authors didn't give up. They realized that if they changed the rules slightly—by looking at the string at a specific "zoom level" (frequency) while shrinking the ball size—they could force the math to work.

• The Metaphor: It's like trying to hear a whisper in a hurricane. You can't hear it directly. But if you build a special microphone that filters out the wind at the exact moment you lean in, you might catch the whisper.
• The Result: By carefully coordinating how small the ball is and how "zoomed in" they look, they could recover the correct energy formula for specific pairs of shapes.

Summary

• Goal: To find a mathematical way to describe the "most likely shape" of a quantum field.
• Method: Using "tiny ball" probability ratios (Onsager-Machlup functionals).
• 1D: Works perfectly.
• 2D: Works if you measure the "roughness" of the shape, not just the shape itself.
• 3D: Fails with standard methods because the field is too chaotic. The "smooth" shapes are impossible for the field to be.
• Takeaway: In the most complex dimensions of our universe (3D), the simple idea of a "probability density" (like a smooth hill of likelihood) might not exist. The landscape is too jagged and broken for our usual maps to work.

Technical Summary

Here is a detailed technical summary of the paper "On the Onsager-Machlup functional of the $\Phi^4_d$ measure" by Ioannis Gasteratos and Zachary Selk.

1. Problem Statement

The paper investigates the existence and form of Onsager-Machlup (OM) functionals for the $\Phi^4_d$ measures (and more generally $P(\Phi)_2$ measures) in finite volume for dimensions $d=1, 2, 3$.

• Context: In Constructive Quantum Field Theory (CQFT), the $\Phi^4_d$ measure is formally defined as a Gibbs measure $\mu \propto e^{-S(\phi)} d\phi$, where $S(\phi)$ is the action functional containing a quartic interaction term. However, in dimensions $d \geq 2$, the field $\phi$ is a distribution, making the product $\phi^4$ ill-defined without renormalization.
• The Gap: While these measures have been rigorously constructed as limits of renormalized approximations (often via Stochastic PDEs), an intrinsic description of their "density" relative to a reference measure (like the Gaussian Free Field, GFF) is often missing or singular.
• Objective: The authors aim to determine if the OM functional—a generalization of the probability density function for infinite-dimensional spaces defined via small ball probability ratios—coincides with the formal action $S(\phi)$. Specifically, they ask: Does $\lim_{r \to 0} \frac{\mu(B_r(z_1))}{\mu(B_r(z_2))} = \exp(S(z_2) - S(z_1))$ hold?

2. Methodology

The authors analyze the OM functional by examining the ratio of probabilities of small metric balls centered at different points $z_1, z_2$. The core methodology involves:

1. Metric Sensitivity: Recognizing that OM functionals are highly sensitive to the choice of topology/metric. Standard metrics often fail in singular SPDE contexts.
2. Enhanced Metrics (Rough Path Analogy): Inspired by Rough Path theory, the authors propose "enhanced" balls. Instead of just controlling the distance $\|\phi - z\|$, these sets control higher-order terms (Wick powers) of the field.
   • For $d=2$, they define balls controlling $\|\phi - z\|$, $\|(\phi-z)^{:2:}\|$, etc.
   • For $d=3$, they attempt to control Wick powers up to the cube, including logarithmic divergence terms.
3. Renormalization and Approximations:
   • They utilize the sequence of renormalized measures $\mu_n$ (Fourier truncated) converging to the physical measure $\mu$.
   • They apply the Cameron-Martin theorem to shift the measure from the center of the ball to the origin, expanding the interaction terms using binomial expansions of Wick powers.
4. Asymptotic Analysis: They analyze the limits as the ball radius $r \to 0$ and, for $d=3$, as the frequency cutoff $n \to \infty$ (often coupled with $r$).

3. Key Contributions and Results

Case $d=1$: Standard OM Functional

• Result: The standard OM functional exists and coincides exactly with the action $S(\phi)$.
• Detail: In one dimension, the field is continuous, and no renormalization is needed. The authors prove that for $z \in H^1_0$, the ratio of probabilities converges to $\exp(S(z_2) - S(z_1))$.
• Significance: This serves as a baseline, confirming that the OM formalism recovers the classical action when the measure is well-behaved.

Case $d=2$: Enhanced OM Functional

• Challenge: In 2D, the measure is absolutely continuous with respect to the GFF, but the density $e^{-:P:(\phi)}$ is not continuous in the standard topology. Small balls in standard norms do not control the Wick powers necessary for the expansion.
• Solution: The authors introduce "enhanced balls" $B_r(z)$ that constrain the field and its Wick powers (e.g., $\|(\phi-z)^{:2:}\| < r$). This mimics the "enhancement" in Rough Path theory where iterated integrals are added to the path to restore continuity.
• Result: With these enhanced sets, the OM functional is well-defined and coincides with the action $S(\phi)$ (including the Wick-renormalized interaction).
• Significance: This bridges the gap between Rough Path analysis and Constructive QFT, showing that continuity of the density can be recovered by enriching the topology.

Case $d=3$: Degeneracy and Joint Limits

• Challenge: In 3D, the $\Phi^4_3$ measure is mutually singular with respect to the GFF and any smooth shift. Furthermore, the Wick cube $\phi^{:3:}$ diverges logarithmically.
• Result 1 (Degeneracy): The authors prove that for any "reasonable" enhanced sets (controlling Wick powers and logarithmic divergences), the OM functional is degenerate. Specifically, for any smooth $z \neq 0$, the probability of the ball $B_r(z)$ is zero (or the ratio diverges to 0 or $\infty$) compared to the ball at $z=0$.
  • Theorem 1.3: The OM functional is effectively $\infty$ for $z \neq 0$ and $0$for$z=0$.
• Result 2 (Restoring Finiteness): Despite the degeneracy, the authors recover the action $S(\phi)$ by considering a joint limit where the ball radius $r \to 0$ and the frequency cutoff $n \to \infty$ simultaneously, subject to the condition $\lim_{r \to 0} r \log n(r) = 0$.
  • Under specific compatibility conditions on $z_1, z_2$ (e.g., matching $L^2$ norms of their truncated versions), the ratio converges to $\exp(S(z_2) - S(z_1))$.
  • They also construct a "third-order" OM ratio (involving 4 points) that cancels out the quadratic divergences, yielding a finite limit for all smooth $z_1, z_2$.

4. Significance and Implications

1. Intrinsic Description of QFT Measures: The work provides a rigorous attempt to define the "density" of singular QFT measures without relying solely on the GFF reference, which fails in $d=3$.
2. Rough Analysis Connection: The $d=2$ result validates the intuition that the "roughness" of the field requires an enhanced topology (similar to Rough Paths) to define meaningful densities.
3. Limitations of OM in $d=3$: The degeneracy result in $d=3$ highlights a fundamental difference between OM functionals and Large Deviation Principles (LDPs). While the LDP rate function for $\Phi^4_3$ is well-defined and equals the action, the OM functional (which relies on local metric properties) is extremely sensitive to the metric choice and often fails to exist in a non-degenerate form for singular measures.
4. Future Directions: The paper suggests that recovering the action in $d=3$ may require non-Gaussian reference measures (as in the variational approach of Barashkov and Gubinelli) or more sophisticated drift-based frameworks, rather than just refining the metric on the space of distributions.

Summary Conclusion

The paper establishes that the Onsager-Machlup functional successfully recovers the $\Phi^4$ action in $d=1$ and $d=2$ (with enhanced metrics), but fails to do so in a standard sense for $d=3$ due to mutual singularity and logarithmic divergences. However, by coupling the small-ball limit with the high-frequency limit, the action can be recovered asymptotically, offering a nuanced view of how "density" behaves in singular quantum field theories.