Moments in the CFT Landscape

This paper introduces a novel numerical bootstrap framework based on moment observables that provides rigorous bounds on conformal field theory spectra, revealing new continuous families of kinks and spectral reorganizations across dimensions $2 < d < 6$ that are difficult to access with traditional methods.

The Gist

Imagine you are trying to understand a massive, chaotic orchestra playing a symphony you've never heard before. You can't see the musicians, and you can't hear every single note clearly because the sound is so dense and complex.

The Old Way (Traditional Bootstrap):
Traditionally, physicists trying to understand these "orchestras" (which are actually Quantum Field Theories describing the universe) have tried to identify specific, famous soloists. They ask: "Who is the first violinist? What is their name? How loud are they?" They look for specific, low-energy notes (light particles) to figure out the rules of the music. This works great for simple songs, but if the orchestra is playing a massive, heavy, complex symphony (with heavy particles), trying to pick out individual soloists becomes impossible. The music is too dense.

The New Way (The Moment Bootstrap):
In this paper, the authors introduce a new, clever way to listen. Instead of trying to identify every single musician, they decide to measure the average properties of the whole orchestra.

Think of it like this:

• Instead of asking "Who is the first violinist?", they ask, "What is the average pitch of the entire orchestra?"
• Instead of asking "How loud is the drummer?", they ask, "What is the average volume?"
• They calculate "moments," which are just fancy statistical averages (like the mean, the spread, or the skew) of all the notes being played.

Why is this cool?

1. It works for heavy music: When the music gets very "heavy" (in physics terms, when the particles involved have high energy), the old method breaks down. It's like trying to count individual grains of sand in a sandstorm. The new "Moment" method is like measuring the density of the sandstorm. It works perfectly even when the music is too loud and complex to pick out individual notes.
2. It finds hidden landmarks: When the authors mapped out these averages, they didn't just get a smooth line. They found sharp corners and cliffs in the map.
   • The "Kinks": Imagine driving a car along a road. Usually, the road is smooth. But suddenly, you hit a sharp turn or a cliff edge. In physics, these sharp turns are called "kinks." They usually mean you've stumbled upon a very special, unique theory of the universe (like the famous Ising model, which describes how magnets work).
   • The Surprise: The authors found new cliffs and valleys that no one knew existed before. These are like hidden valleys in a mountain range that no hiker has ever seen.

The "Fake" Ghosts:
One of the most fascinating discoveries is something the authors call the "Fake-Primary" effect.
Imagine you are looking at a shadow on a wall. Sometimes, the shadow looks like a person, but it's actually just a trick of the light caused by a different object behind it.
In their math, they found that at certain points, the "ghost" of a particle (a mathematical shadow) looks exactly like a real particle. The math gets confused and thinks a particle exists when it's actually just a mathematical artifact. The authors figured out how to "remap" these ghosts to understand what's really going on. It's like realizing the scary monster in the closet is just a coat rack, and then using that knowledge to understand the room better.

The "Cliff" and the "Valley":
The authors describe the landscape of these theories as a terrain:

• The Cliff: A sudden drop where a specific type of particle suddenly disappears from the orchestra.
• The Valley: A low point where the music settles into a specific, calm pattern.
• The Hill: A peak where the music gets chaotic and unbounded.

The Big Picture:
This paper is essentially a new GPS for the universe's rulebook.

• The old GPS only worked well for simple, quiet neighborhoods (light particles).
• This new GPS works for the whole city, including the busy, noisy downtown areas (heavy particles).
• It has discovered new neighborhoods (the "kinks" and "valleys") that we didn't know existed.
• It helps us understand that even when we can't see the individual players, we can still understand the whole song by listening to the average rhythm.

In short: The authors built a new mathematical tool that listens to the "average voice" of the universe's particles instead of shouting at individual ones. This allowed them to find new, hidden structures in the laws of physics that were previously invisible, proving that sometimes, looking at the big picture is better than staring at the details.

Technical Summary

1. Problem Statement

The numerical conformal bootstrap has traditionally focused on isolating specific Conformal Field Theories (CFTs) by optimizing functionals that target low-lying operator properties, such as the scaling dimension of the first non-identity operator (gap maximization) or the OPE coefficient of a specific operator. While highly successful for light correlators (small external scaling dimensions $\Delta_\phi$), these methods face significant challenges in the heavy correlator regime (large $\Delta_\phi$). In this regime, the spectrum becomes dense, and standard gap-maximization functionals converge extremely slowly or fail to produce meaningful bounds. Furthermore, traditional approaches often miss global, collective features of the operator spectrum that are not captured by focusing on individual operators.

The authors ask: Can one define and rigorously bound global observables of the Operator Product Expansion (OPE) spectrum that remain effective in both light and heavy regimes, and what new geometric structures do these observables reveal in the space of CFTs?

2. Methodology: The Moment Bootstrap

The authors introduce the numerical moment bootstrap, a framework that replaces the optimization of individual operator properties with the optimization of OPE moments.

• Definition of Moments: Instead of bounding a single $\Delta$ or $\lambda$, the authors define moments as weighted averages over the conformal data. For a four-point function of identical scalars, the spectral density is defined at the self-dual point ($z=\bar{z}=1/2$) as:
  $$p_\ell(\Delta) = \sum_O \lambda_{\phi\phi O}^2 g_{\Delta, \ell}(1/2, 1/2) \delta(\Delta - \Delta_O) \delta_{\ell, \ell_O}$$
  The moments are then defined as:
  $$\nu_{m,n} = \sum_\ell \int d\Delta \, p_\ell(\Delta) \, \Delta^m \ell^n$$
  Normalized moments (e.g., $\langle \Delta^k \rangle = \nu_k / \nu_0$) are used to remove dependence on the overall normalization of the correlator.

• Numerical Implementation:

  • The problem is cast as an infinite-dimensional linear program subject to crossing symmetry and unitarity constraints.
  • Using the rational approximation of conformal blocks (expanding in radial coordinates), the crossing equations are converted into polynomial constraints on the scaling dimension $\Delta$.
  • The problem is solved using Semidefinite Programming (SDP) via the solver SDPB. The dual formulation involves finding a functional (a linear combination of derivatives of conformal blocks) that is non-negative over the allowed spectrum and minimizes the moment value.
  • Spectrum Extraction: At the optimal solution, the extremal spectrum is extracted using complementary slackness conditions, allowing for the reconstruction of the spectral density.

• Heavy Limit Analysis:

  • The authors utilize Maximum Entropy (Max-Entropy) reconstruction to approximate the continuous spectral density from the finite set of computed moments. This assumes the OPE spectrum follows a distribution that maximizes Shannon entropy given the known moments, effectively capturing the coarse-grained behavior of the heavy limit.

3. Key Contributions

1. Novel Framework: The development of a rigorous numerical bootstrap for moment variables, shifting the focus from individual operators to global spectral averages.
2. Heavy Regime Success: Demonstrating that moment bounds converge rapidly to analytically derived power laws in the heavy correlator limit ($\Delta_\phi \to \infty$), a regime where traditional gap-maximization fails.
3. Discovery of New Kinks: The identification of two continuous families of kinks in the lower bounds of moment variables across spatial dimensions $2 < d < 6$. These kinks persist where traditional methods see smooth boundaries.
4. Unified View of Ising Physics: Showing that the 3d Ising model kink appears naturally in the space of $\Delta$-moments and $\ell$-moments, reproducing known results while providing a new geometric perspective.
5. Spectral Reorganization: Detailed analysis of the "fake-primary effect" and operator decoupling phenomena that drive the geometric features of the moment landscape.

4. Key Results

A. Heavy Correlator Limit

• In the limit of large external dimension $\Delta_\phi$, the numerical moment bounds converge rapidly to the analytic bounds derived in [36]:
  $$2^{k/2} \leq \lim_{\Delta_\phi \to \infty} \frac{\langle \Delta^k \rangle}{\Delta_\phi^k} \leq 2^{3k/2 - 1}$$
• The bounds are saturated by Generalized Free Field (GFF) correlators.
• Comparison: Moment maximization converges to the linear trajectory $\sqrt{2}\Delta_\phi$ even at low derivative orders ($\Lambda \approx 3$), whereas gap maximization fails to produce bounds at large $\Delta_\phi$ for the same order.

B. The Ising Model

• The 3d Ising model appears as a sharp kink on the upper bound of the first normalized moment $\langle \Delta \rangle_*$ (excluding the identity) at $\Delta_\phi \approx 0.518$.
• This kink is also visible in bounds on the spin moment $\langle \ell \rangle_*$, reflecting the extremization of the stress tensor coupling (central charge minimization) and the decoupling of operators in the Wilson-Fisher theory.

C. The "Moment Landscape" (Lower Bounds)

The most significant discovery lies in the lower bounds of the moment variables, which reveal a rich geometric structure across $2 < d < 6$:

1. Two Families of Kinks:
   • The "Cliff": A sharp drop in the moment bound associated with the decoupling of the second-lightest scalar operator. This operator's dimension is close to $2\Delta_\phi$ (the GFF double-trace value). This family connects continuously to the free scalar theory at $d=6$.
   • The "Second Valley": A second sharp feature where the lightest scalar operator saturates the unitarity bound ($\Delta = (d-2)/2$).
2. Spectral Features:
   • Operator Decoupling: The kinks correspond to specific events where operators drop out of the extremal spectrum (OPE coefficient $\to 0$).
   • Fake-Primary Effect: At the second valley, the lightest scalar hits the unitarity bound. The authors show that the solution can be "remapped" to its shadow (a "fake primary") at $\Delta = (d+2)/2$. This explains why the lower bound can be saturated by solutions that appear to violate standard gap assumptions but are consistent with crossing symmetry due to the divergence of the conformal block at the unitarity bound.
   • Unbounded Correlators: In the region between the "cliff" and the "second valley" (the "hill"), the reconstructed correlators become unbounded from above, suggesting solutions where the identity operator is negligible.

D. Dimensional Dependence

• $d=2$: The lower bound is smooth and linear ($\sqrt{2}\Delta_\phi$), lacking the kinks seen in higher dimensions.
• $2 < d < 6$: The landscape exhibits the "Cliff $\to$ Valley $\to$ Hill $\to$ Valley" structure.
• $d \to 2^+$: The features merge into a single kink, and the behavior becomes dominated by the fake-primary effect.

5. Significance and Future Directions

• New Probe of Theory Space: The moment bootstrap provides a "coarse-grained" probe of the CFT landscape, revealing collective structures and spectral reorganizations that are invisible to traditional gap-maximization techniques.
• Bridge Between Numerics and Analytics: It successfully bridges the gap between numerical SDP methods and analytic heavy-limit results, offering a unified framework.
• Physical Interpretation: The discovery of continuous families of kinks suggests the existence of new, previously unexplored solutions to crossing symmetry, potentially related to renormalization group flows or exotic sectors of CFTs.
• Future Work: The authors suggest extending this to mixed correlators (heavy-light, heavy-heavy) to probe finite-temperature physics and using mixed-correlator constraints to isolate these new kink solutions into "bootstrap islands."

In summary, this paper establishes the moment bootstrap as a powerful new tool that not only reproduces known results (like the Ising kink) with high precision but also uncovers a complex, previously hidden geometric structure in the space of CFTs, driven by operator decoupling and the fake-primary effect.