Mapping the CFT Landscape with Moment Constraints

Author: Denis Avetisyan

A new technique leverages semidefinite programming to chart the space of conformal field theories by analyzing moments of operator product expansion data.

The landscape of lightest scalar operators in conformal field theory exhibits distinct regions-a “pond” devoid of light scalars, and areas where operators like φ drift toward the unitarity bound or follow the <span class="katex-eq" data-katex-display="false">2\Delta_{\phi}</span> trajectory-separated by boundaries indicative of operator decoupling, suggesting a fundamental structure despite the absence of a perturbative explanation for these observed behaviors. — The landscape of lightest scalar operators in conformal field theory exhibits distinct regions-a “pond” devoid of light scalars, and areas where operators like φ drift toward the unitarity bound or follow the $2\Delta_{\phi}$ trajectory-separated by boundaries indicative of operator decoupling, suggesting a fundamental structure despite the absence of a perturbative explanation for these observed behaviors.

This review details a ‘moment bootstrap’ approach using crossing symmetry and spectral density bounds to explore previously uncharted regions of the CFT landscape.

Despite ongoing progress in the conformal bootstrap, fully characterizing the landscape of consistent quantum field theories remains a formidable challenge. This paper, ‘Moments in the CFT Landscape’, introduces a novel numerical approach-the ‘moment bootstrap’-which leverages semidefinite programming to bound moments of operator product expansion data and map out the space of conformal field theories. These moment variables reveal previously hidden geometric structures and continuous families of solutions associated with nontrivial spectral reorganizations, offering a complementary perspective to traditional gap maximization techniques. Could this moment-based approach unlock a more complete understanding of the organizing principles governing the vast space of possible quantum field theories?

The Algorithmic Foundation of Conformal Field Theory

Conformal Field Theories, or CFTs, represent a cornerstone in theoretical physics, appearing in models ranging from condensed matter systems and cosmology to string theory and quantum gravity. Despite their pervasive influence, obtaining exact, analytical solutions for CFTs proves remarkably difficult – a challenge stemming from the theory’s inherent complexity and infinite degrees of freedom. This scarcity of solutions motivates the development of non-perturbative methods, techniques that don’t rely on approximations valid only for weakly interacting systems. These approaches aim to circumvent the limitations of traditional perturbation theory and explore the full spectrum of possible CFTs, unlocking a deeper understanding of their behavior and providing insights into the fundamental laws governing the universe. The inability to solve CFTs analytically highlights the need for innovative tools capable of navigating the intricate landscape of quantum field theory.

Conformal Field Theories, while central to areas ranging from condensed matter physics to string theory, are hampered by an inherent complexity arising from their infinite number of possible operators – functions describing the system’s behavior. Unlike theories with a finite set of relevant parameters, CFTs present an infinite-dimensional space of possibilities, making traditional perturbative methods-which rely on expanding around a known solution-inadequate. This vastness poses a significant challenge; standard analytical techniques struggle to navigate and extract meaningful information from this landscape. Consequently, physicists have been compelled to develop innovative, non-perturbative tools capable of exploring this infinite-dimensional space without requiring a pre-defined starting point, ultimately seeking to constrain and characterize the allowed configurations within these complex theories.

The Conformal Bootstrap represents a paradigm shift in the study of Conformal Field Theories by sidestepping the traditional requirement for explicit mathematical solutions. Instead of directly solving for a theory’s properties, this numerical approach harnesses the power of symmetry and consistency conditions. It operates by imposing constraints on the space of possible CFTs, effectively narrowing down the landscape of viable theories without needing to know the exact details of their constituent parts. This is achieved through a sophisticated process of bootstrapping – starting with basic axioms and iteratively refining the theory based on self-consistency – allowing researchers to explore a vast range of CFTs and uncover previously inaccessible phenomena. The method’s strength lies in its ability to reveal universal properties and critical exponents with high precision, even in scenarios where analytical techniques fail, offering a potent tool for understanding diverse physical systems from condensed matter physics to cosmology.

Gap maximization exhibits significantly slower convergence and fails to establish bounds at high scaling dimensions, unlike moment maximization-which rapidly converges to <span class="katex-eq" data-katex-display="false">\sqrt{2}\Delta_{\phi}</span> and accurately reproduces asymptotic scaling at low derivative orders. — Gap maximization exhibits significantly slower convergence and fails to establish bounds at high scaling dimensions, unlike moment maximization-which rapidly converges to $\sqrt{2}\Delta_{\phi}$ and accurately reproduces asymptotic scaling at low derivative orders.

Precision Tools for Mapping the Operator Landscape

The Operator Product Expansion (OPE) is a fundamental tool in conformal field theory and related areas, providing a systematic way to analyze the short-distance behavior of local operators $\mathcal{O}_i$ and $\mathcal{O}_j$ . The OPE postulates that the product of two local operators can be expressed as a sum over a complete set of local operators $\mathcal{O}_k$ , with each term containing a coefficient $C_{ijk}$ that encapsulates the strength of the interaction and a specific operator product structure. Formally, this is represented as $\mathcal{O}_i(x) \mathcal{O}_j(0) = \sum_k C_{ijk} \partial^n \mathcal{O}_k(0)$ , where the sum is over all local operators and $n$ represents derivatives acting on $\mathcal{O}_k$ . This expansion allows the calculation of correlation functions, such as $\langle \mathcal{O}_1(x_1) \mathcal{O}_2(x_2) \dots \rangle$ , by replacing operator products with sums over OPE coefficients and local operators, effectively reducing complex multi-operator expressions to simpler, more manageable forms.

Semidefinite Programming (SDP) is utilized within the bootstrap framework to address the linear constraints that arise when analyzing operator product expansions. These constraints, derived from requirements like unitarity and crossing symmetry, define a feasible region for the space of conformal blocks. SDP provides a computationally efficient method for finding solutions within this region and, crucially, for establishing lower bounds on relevant quantities such as the dimensions of operators. By relaxing the non-negativity constraints on certain quantities, SDP transforms the original problem into a tractable convex optimization task, solvable using established algorithms. The resulting bounds, while not necessarily exact, provide valuable information about the allowed spectrum of operators and can be iteratively refined to improve accuracy. Δ and $L$ are commonly bounded using SDP techniques.

The Lorentzian Inversion Formula provides a method for determining a real-valued correlation function $G(x)$ given its imaginary part $Im(G(x))$ . This is achieved through an integral transformation involving the Hilbert transform, expressed as $G(x) = \frac{1}{\pi} \in t_{-\in fty}^{\in fty} \frac{Im(G(k))}{k-x} dk$ . Computationally, this formula is advantageous because calculating the imaginary part of the correlation function is often simpler than directly computing the real part, especially in scenarios involving dispersion relations or high-energy behavior. By leveraging this simplification, the Lorentzian Inversion Formula enables efficient reconstruction of the complete correlation function from readily obtainable data, reducing computational complexity in various theoretical and numerical investigations.

Maximum-entropy reconstruction accurately approximates the discrete measure of the <span class="katex-eq" data-katex-display="false">\langle\phi\phi\phi\phi\rangle</span> correlator across varying external scaling dimensions <span class="katex-eq" data-katex-display="false">\Delta_\phi</span>. — Maximum-entropy reconstruction accurately approximates the discrete measure of the $\langle\phi\phi\phi\phi\rangle$ correlator across varying external scaling dimensions $\Delta_\phi$ .

Decoding the Operator Spectrum: A Landscape of Dimensions and Spins

The Conformal Bootstrap method characterizes the spectrum of operators within a Conformal Field Theory (CFT) by leveraging data derived from the Operator Product Expansion (OPE). Specifically, the method focuses on the moments of the OPE coefficients, which encode information about the scaling dimensions Δ and spins $s$ of all primary operators. By systematically analyzing these moments, the bootstrap constructs a set of constraints that define the allowed spectrum. The resulting description is considered coarse-grained as it primarily focuses on identifying the leading behavior of correlation functions and does not necessarily resolve all details of the operator spectrum, but it provides a powerful tool for studying the global properties of the CFT and identifying universal features.

Unitarity, a fundamental requirement of quantum field theory, dictates that probabilities must be positive and real, thereby constraining the allowed operator dimensions and ensuring a physically consistent theory. The implementation of unitarity conditions within the Conformal Bootstrap program significantly reduces the solution space and stabilizes numerical calculations. Complementary to unitarity is the Gap Assumption, which posits a minimal dimension $\Delta_{min}$ for the lowest non-identity primary operator. This assumption, while not rigorously proven for all conformal field theories, serves a crucial practical role by accelerating the convergence of numerical algorithms used to solve the bootstrap equations; without it, the computational demands increase substantially, hindering the ability to efficiently explore the operator spectrum and characterize the CFT.

Analysis of the spectral density, which maps the distribution of operator dimensions and spins, has revealed the existence of two continuous families of kinks observed across the dimension range of 2 < d < 6. These kinks represent previously unidentified geometric structures indicative of spectral reorganization within the Conformal Field Theory. The location and characteristics of these kinks are directly related to the CFT’s underlying structure and can provide critical information regarding phase transitions and the behavior of the theory at different energy scales. Further investigation into these spectral features may elucidate connections between operator dimensions, spins, and the geometry of the CFT’s operator space.

Analysis of the first normalized moment, computed at <span class="katex-eq" data-katex-display="false">\Lambda=23</span>, reveals bounds on the allowed moment space-indicated by shaded regions that darken with stronger gap assumptions-and confirms consistency with moments extracted from the Ising model and solutions from the Gaussian Free Field, demonstrating that the upper bounds remain consistent regardless of the gap choice. — Analysis of the first normalized moment, computed at $\Lambda=23$ , reveals bounds on the allowed moment space-indicated by shaded regions that darken with stronger gap assumptions-and confirms consistency with moments extracted from the Ising model and solutions from the Gaussian Free Field, demonstrating that the upper bounds remain consistent regardless of the gap choice.

Extending the Reach: Validation, Limits, and Impact

The conformal bootstrap method isn’t limited to simple scenarios; it extends remarkably into the “Heavy Correlator Regime”. This regime is characterized by a proliferation of operator contributions – a situation where an increasingly dense set of possible interactions must be considered. While mathematically elegant, this density dramatically increases the computational demands, pushing the limits of current numerical techniques. Effectively, the method is tackling problems where the sheer number of variables and relationships strains even powerful computing resources, yet continues to yield meaningful results by systematically exploring the space of possible solutions and identifying consistent theories. This capability demonstrates the bootstrap’s robustness and potential for uncovering subtle physics in complex systems, even as computational feasibility remains a significant challenge.

The conformal bootstrap’s efficacy hinges on its ability to reproduce known physics, and rigorous benchmarking against established models provides crucial validation. Researchers frequently test the bootstrap’s predictions using the Ising Model, a well-understood system exhibiting a phase transition, as a control case. By successfully recovering known critical exponents and operator dimensions for the Ising Model, the bootstrap demonstrates its internal consistency and reliability in exploring more complex, less-understood systems. This process confirms that the numerical methods employed are robust and yield physically meaningful results, bolstering confidence in the bootstrap’s application to a wider range of theoretical investigations – ultimately establishing it as a powerful, non-perturbative tool in quantum field theory.

Investigations frequently initiate with a Generalized Free Field, a mathematically tractable system that serves as a foundation for tackling more intricate scenarios. Through this approach, researchers have pinpointed a significant decrease in the moment variable occurring at an external scaling dimension – denoted as $\Delta\phi$ – of approximately 0.908. This abrupt drop signifies a phenomenon known as operator decoupling, where certain complex interactions effectively ‘switch off’, simplifying the overall system and providing crucial insights into its underlying structure. The precision of this decoupling point serves as a valuable benchmark, allowing scientists to validate the accuracy of their calculations and explore the limits of this powerful bootstrap technique in understanding quantum field theories.

Correlator bounds, calculated at <span class="katex-eq" data-katex-display="false">\Lambda = 11</span>, converge at the free theory limit (sharp left edge) and include the 3d Ising correlator (red dot) as determined by conformal data from Simmons-Duffin:2016wlq, consistent with the results of Paulos:2021jxx. — Correlator bounds, calculated at $\Lambda = 11$ , converge at the free theory limit (sharp left edge) and include the 3d Ising correlator (red dot) as determined by conformal data from Simmons-Duffin:2016wlq, consistent with the results of Paulos:2021jxx.

Refinements and Future Directions: Pushing the Boundaries

The Conformal Bootstrap method incorporates a sophisticated mechanism for distinguishing genuine primary operators from ‘fake primaries’ – operators that superficially resemble primaries but are, in fact, descendants of other operators. This differentiation is crucial for obtaining accurate results, as including these ‘fake’ contributions would skew calculations of physical properties. The method achieves this by meticulously examining the operator product expansion (OPE) and correlation functions, identifying inconsistencies that reveal the descendant nature of these seemingly primary fields. By effectively filtering out these spurious operators, the Conformal Bootstrap ensures that only physically relevant contributions are considered, leading to a more reliable and precise understanding of conformal field theories and the systems they describe – from critical phenomena to condensed matter physics. Δ values are carefully analyzed to confirm the identification and exclusion of these operators.

The Conformal Bootstrap’s remarkable versatility allows it to move beyond theoretical simplicity and tackle a diverse array of Conformal Field Theories (CFTs) in any number of spacetime dimensions, from the well-studied two-dimensional systems to those encountered in more complex, higher-dimensional scenarios. This adaptability isn’t merely mathematical; the method provides a powerful toolkit for investigating real-world physical phenomena governed by conformal symmetry, including critical phenomena – the behavior of systems at precise transition points – and the exotic states of matter explored within condensed matter physics. By circumventing the need for pre-defined models, the bootstrap can reveal universal properties and uncover novel phases, offering insights into systems ranging from magnetic materials to high-temperature superconductors, and ultimately broadening the understanding of collective behavior in nature.

The conformal bootstrap’s potential is increasingly realized through ongoing refinements in numerical methodologies and the ever-growing availability of computational resources. Recent studies demonstrate that improvements in these areas allow for more precise calculations of scaling dimensions, notably observing decoupled operator dimensions approaching $2\Delta\phi$ , a strong indicator of the decoupling process’s validity. Simultaneously, reconstructed correlation functions reveal unbounded growth within the ‘hill’ region, suggesting that a relatively small number of low-lying operators dominate the system’s behavior. These advancements not only validate existing theoretical frameworks but also pave the way for exploring more complex systems and potentially uncovering previously inaccessible aspects of fundamental physics, ranging from critical phenomena to condensed matter systems and beyond.

The fake-primary remap, computed at <span class="katex-eq" data-katex-display="false">\Lambda = 23</span>, effectively reconstructs the lower bound on the first moment (ind=3d=3) using a tolerance of <span class="katex-eq" data-katex-display="false">10^{-4}</span>, as demonstrated by the correspondence with the moment bounds in figure 1(b) and detailed in appendix E. — The fake-primary remap, computed at $\Lambda = 23$ , effectively reconstructs the lower bound on the first moment (ind=3d=3) using a tolerance of $10^{-4}$ , as demonstrated by the correspondence with the moment bounds in figure 1(b) and detailed in appendix E.

The pursuit within this paper, mapping the conformal field theory landscape through moment bootstrapping, echoes a fundamental tenet of mathematical rigor. It’s not merely about finding solutions-functional bounds on OPE data-but establishing their consistency and validity. As David Hume observed, “The mind is willing to believe what it wishes.” This sentiment applies subtly; the researchers don’t simply wish for a particular CFT structure, but demonstrate its feasibility through the constraints of crossing symmetry and semidefinite programming. The method’s ability to reveal previously unexplored geometric structures highlights that true understanding stems from provable consistency, rather than empirical observation alone. The extremal functional, central to this approach, serves as a mathematical predicate-a clear condition for the truth of a theoretical construct.

What Lies Beyond?

The presented ‘moment bootstrap’ offers a refinement, not a revolution. It is tempting to declare a new era in conformal field theory exploration, yet the fundamental challenge remains: extracting genuine, analytically-provable results. The technique sidesteps, rather than solves, the inherent difficulty of dealing with infinite-dimensional data. While bounding moments provides constraints, these constraints, absent further mathematical machinery, are merely shadows of the full solution. Optimization without analysis remains self-deception, a trap for the unwary engineer.

Future efforts must address the limitations imposed by semidefinite programming itself. The duality between moment problems and measure theory beckons, but fully exploiting this connection demands overcoming computational bottlenecks and, crucially, developing methods to guarantee the uniqueness of recovered spectral densities. The existence of multiple solutions, each satisfying the imposed moment constraints, would render the entire endeavor aesthetically unsatisfying, a numerical flourish masking a lack of true understanding.

Ultimately, the pursuit of conformal field theories hinges not on generating ever more elaborate datasets, but on discovering the hidden symmetries and mathematical structures that govern their behavior. This moment bootstrap technique, while valuable, is but one tool in that larger, more ambitious undertaking – a step toward rigorous, provable knowledge, not merely empirically-validated approximations.

Original article: https://arxiv.org/pdf/2603.18140.pdf

Contact the author: https://www.linkedin.com/in/avetisyan/