The Principle of Maximum Freedom: Extremizing the Free Energy in Melonic QFTs

TL;DR

Quantum Field Theory (QFT) is notoriously difficult at strong coupling. This thesis reveals that a specific class of theories—melonic large-N theories—can be solved exactly by a principle of "maximum freedom." By extremizing $ildeF$ (the universal part of the sphere free energy), we can determine the scaling dimensions of complex interacting fields without ever performing a traditional Feynman sum.

This work establishes a deep theoretical bridge between Large-N Melonic theories and Supersymmetric CFTs, proving they share the same infrared optimization logic.

Problem & Motivation: Beyond the Vector/Matrix Dichotomy

In the landscape of solvable large-N limits, we usually have two extremes:

1. Vector Models ($N$ components): Summing "cactus" diagrams. These are simple but often lead to "mean field" results that don't capture the full richness of interaction.
2. Matrix Models ($NimesN$ components): Summing "planar" diagrams. These are rich enough to describe Quantum Gravity but are generally impossible to sum analytically in higher dimensions.

Melonic theories (like the SYK model and Tensor models) sit in the "Goldilocks zone." They are dominated by melonic graphs—a subset of planar graphs that are sufficiently complex to be interesting but simple enough to be resummable. The challenge is finding a universal "master principle" to solve them across any dimension $d$.

Methodology: The $ildeF$-Extremization Principle

The core insight of the thesis is that the infrared (IR) behavior of these theories is governed by the universal part of the sphere free energy, denoted as $ildeF$.

1. The 2PI Physics

Using the Two-Particle-Irreducible (2PI) formalism, the author shows that the free energy of a melonic QFT can be written as an effective action of the propagator. In the melonic limit, this action simplifies to: $\frac{1}{N}ildeF(\{{\Delta }_{\varphi }\})={\sum }_{fields}ilde{F}_{GFF}({\Delta }_{\varphi })+extConstraints$ where $ilde{F}_{GFF}$ is the free energy of a Generalized Free Field.

2. Constraints as Lagrange Multipliers

Interactions in the Lagrangian (like ${\varphi }^2\bar{\psi }\psi$) don't just change the numbers; they impose marginality constraints. For an interaction to survive in the IR, its scaling dimension must sum to $d$.

The author proves that the running coupling constants effectively act as Lagrange multipliers enforcing these constraints. Therefore, finding the IR physics is equivalent to:

• Extremizing the sum of individual field free energies.
• Subject to the linear constraints of the interaction.

Figure 1: (User Note: Replace with Figure 4.11 from paper) The spectrum of bilinears shows windows of stability where scaling dimensions are real, separated by regions where operators complexify.

The Case Study: The Quartic Yukawa Model

The author applies this to the Quartic Yukawa Model (${\varphi }^2\bar{\psi }\psi +{\varphi }^6$). This model is particularly rich because it contains both bosons and fermions.

Key findings include:

• Multiple Vacua: Unlike simpler models, this theory has several fixed points (melonic, prismatic, and fermionic).
• Supersymmetry Alignment: At a specific ratio of bosonic to fermionic degrees of freedom ($r=2$), the melonic solution perfectly reproduces the result of a supersymmetric Wess-Zumino model.
• Conservation Laws: The spectrum naturally contains the Stress-Energy tensor ($\Delta =d,s=2$) and conserved currents, validating the conformal nature of the solution.

Figure 2: (User Note: Replace with Figure 4.10 from paper) Evolution of scaling dimensions across dimensions $d$, showing the collision and complexification of fixed points.

Deep Insight: Stability and Unitarity

One of the most striking aspects of the work is the discovery of Windows of Stability. In some dimensions, the scaling dimensions are real; in others, they become complex.

In the language of AdS/CFT, complex scaling dimensions correspond to tachyonic instabilities (violations of the Breitenlohner-Freedman bound). This tells us that not all "mathematical" solutions to the melonic equations are "physical" vacua—only those where $ildeF$ is maximized/minimized within certain bounds are stable candidates for the IR.

Critical Analysis & Conclusion

Takeaway

The success of $ildeF$-extremization suggests that the large-N limit "cleans" the theory of its microscopic details, leaving only the most efficient arrangement of degrees of freedom. This provides a prescriptive tool: if you know your field content and your interaction degree, you can solve the IR CFT using nothing more than Gamma functions.

Limitations

• Beyond Large-N: The linear nature of the constraints is a byproduct of the melonic limit. At $O(1/N)$, the "Lagrange multiplier" interpretation likely becomes more complex.
• Vacuum Selection: While we can find the extrema, the theory doesn't yet uniquely predict which extremum the RG flow will land on without further information.

Future Work

The author points towards investigating the Bulk Duals of these theories. The huge number of gauge-invariant operators in melonic theories suggests a gravity dual much more complex than standard AdS/CFT, possibly involving stringy degrees of freedom emerging from the tensor indices.