Conformal Bootstrap Unblocked

summary

The gist

We develop a novel conformal bootstrap formulated directly in terms of conformal symmetry, reflection positivity, and crossing symmetry…

In short

This work develops a novel conformal bootstrap using Gram matrices derived from local operator states, bypassing explicit operator product expansions or conformal blocks. By imposing constraints from conformal symmetry, reflection positivity, and crossing symmetry through a hierarchy of semidefinite programs (SDPs), the method yields bounds on correlation functions. It successfully reproduces results for critical Ising kinks in 2D and 3D and shows that joint positivity significantly tightens bounds.

Key concepts

Gram Matrix Bootstrap
This approach uses Gram matrices, which are positive semidefinite by reflection positivity, to directly constrain correlation functions. By constructing these matrices from states created by pairs of local operators and imposing spectral constraints via conformal generators, the method derives bounds without needing explicit conformal blocks.
Crossing Symmetry
Crossing symmetry is a fundamental property of scattering amplitudes that relates different kinematic configurations. In this bootstrap, it is used to constrain the elements of the Gram matrix, allowing for global consistency checks on correlation functions across various kinematic points.
Semidefinite Programs (SDPs)
The method generates a hierarchy of SDPs where constraints are systematically strengthened by enlarging the space of probe states. Feasibility is tested by maximizing an auxiliary variable subject to positivity conditions, allowing for the determination of gap maximization bounds and certifying excluded gaps.

Terminology used across episodes

This episode discusses

The paper

Conformal Bootstrap Unblocked · Read on arXiv

Li-Yuan Chiang, David Poland

Yale University

We develop a novel conformal bootstrap formulated directly in terms of conformal symmetry, reflection positivity, and crossing symmetry, without explicitly relying on the operator product expansion or conformal blocks. By constructing Gram matrices from states created by pairs of local operators and imposing spectral constraints through conformal generators, we obtain a hierarchy of semidefinite programs that directly constrain correlation functions. We demonstrate the method in one to three dimensions, obtaining correlator and spectral bounds that capture generalized free field behavior and reproduce the characteristic Ising kinks in two and three dimensions. We further show that joint positivity of states at different insertion configurations naturally constrains correlators away from self-dual kinematics.

Transcript

Introduction to the show: ident: High Energy Physics: Theory. The day in high-energy theory: gravity, strings, fields and the questions left open.

Kai: We're looking at the paper: "Conformal Bootstrap Unblocked".

Mira: Here's the gist of the paper. We develop a novel conformal bootstrap formulated directly in terms of conformal symmetry, reflection positivity, and crossing symmetry, without explicitly relying on the operator product expansion or conformal blocks By constructing Gram matrices from states created by pairs of local operators and imposing spectral constraints through conformal generators, we obtain a hierarchy of semidefinite programs that directly constrain correlation functions Core Methodology The approach is based on the elementary observation that any finite collection of states created by local operator insertions defines a positive semidefinite Gram matrix, which is a consequence of reflection positivity Conformal symmetry allows for expressing the elements completely in terms of the four-point function and its derivatives with respect to conformally invariant cross ratios, which are further constrained by crossing The method leads naturally to a hierarchy of semidefinite programs (SDPs) whose constraints can be systematically strengthened by enlarging the space of probe states Dimensional Extensions and Features The construction is demonstrated in one to three dimensions, obtaining correlator and spectral bounds that capture generalized free field behavior and reproduce the characteristic Ising kinks in two and three dimensions In higher dimensions, imposing Bose symmetry via the antipodal configuration allows for the vanishing of odd-spin descendant contributions, which implies that the operator p l(J 2) is positive semidefinite on subspaces containing only even angular momenta Furthermore, joint positivity at different insertion configurations naturally constrains correlators away from self-dual kinematics Constraint Refinement Techniques The hierarchy of SDPs is systematically strengthened by increasing the word length L and including translation generators to construct probe spaces VL To maximize the gap, one can extremize the correlator by solving an SDP subject to Gram matrix positivity conditions and linear crossing equations The antipodal configuration is shown to lead to stronger bounds on both the correlator and the gap compared to asymmetric configurations Application and Convergence The framework successfully reproduces bounds comparable to those obtained from the conventional conformal block bootstrap at similar derivative orders, even though it does not explicitly resolve conformal multiplets The results for the critical Ising models in d = 2 and d = 3 show that imposing Bose symmetry up to spin l = 6 is sufficient to resolve the kinks Away from the self-dual point, joint positivity of Gram matrices significantly tightens bounds compared to decoupled positivity Future Directions The work suggests that the Gram matrix bootstrap can be extended to higher-point correlation functions, where the positivity of states created by multiple operator insertions provides additional consistency conditions Further exploration includes extending the approach to Lorentzian kinematics and investigating connections with moment problems and extremal functional methods The equivalence to the conventional conformal block bootstrap in the infinite-truncation limit is suggested through spectral filtering techniques Computing Gram Matrix Elements The Gram matrix elements are computed by differentiating the overlap of states with respect to insertion coordinates and then converting these derivatives into correlator derivatives with respect to cross ratios via a change of variables The structure of the little group of the antipodal configuration allows for the decomposition of Gram matrices into orthogonal symmetry sectors labeled by helicity and parity SDP Feasibility Problem Feasibility is tested by maximizing an auxiliary variable t subject to the condition G(g) - tI ⪰ 0, which effectively maximizes the smallest eigenvalue of G(g) An alternative formulation involves minimizing Tr(Y G0) subject to Y ⪰ 0, TrY = 1, and Tr(Y Ga) = 0 for all a The bisection method is used to find gap maximization bounds, with excluded gaps certified by dual feasibility Discussion The paper concludes that conformal symmetry, crossing symmetry, and unitarity can be directly utilized in the bootstrap by probing the CFT Hilbert space through Gram matrix positivity The framework offers flexibility in constraining correlators away from the crossing-symmetric point, probing crossing and unitarity more globally Despite its encouraging results, improving numerical efficiency remains a key direction for future work The flexibility of the framework opens several directions for future development, particularly for bootstrap problems where the explicit computation of conformal blocks becomes challenging The paper demonstrates that in d = 3, imposing Bose symmetry constraints up to spin l = 6 is already sufficient to resolve the kinks Computing Gram matrix elements Given the probe space defined in (4.8), the Gram matrix elements can be obtained by differentiating the overlap ⟨Ψ(c', s')Ψ(c, s)⟩ = G(u, v) − 1 with respect to c and s Since conformal covariance fixes the dependence of this overlap on the cross ratios, we can express the matrix elements as G ij (x1, x2, x3, x4) = X m,n A m,n ij (x1, x2, x3, x4)∂m u ∂n v G(u, v) − δi0δj0,(A.1) The Gram matrix can be further simplified by exploiting the little group of the antipodal configuration, which describes the spatial transformations that preserve the insertion points, including their exchange permitted by Bose symmetry B SDP feasibility problem At a fixed external scaling dimension and an assumed spectral gap, our goal is to determine whether there exists a set of correlator derivatives satisfying crossing symmetry and all the Gram matrix positivity constraints Let us write the positivity condition as G(g) = G0 + X a g a G a ⪰ 0, where G0 and Ga are known matrices, and g a are the remaining independent correlator derivatives after crossing is imposed To test feasibility, we introduce an auxiliary variable t and solve Maximize t, subject to G(g) − tI ⪰ 0 An alternative formulation considers the problem from the dual side: Minimize Tr(Y G0), subject to Y ⪰ 0, TrY = 1, Tr(Y Ga) = 0 for all a Any strictly negative objective value then contradicts primal feasibility: Tr

Y G(g): = Tr(Y G0) + X a g aTr(Y G a) = z primal feasibility Where the gains come from The primary gains stem from directly constraining correlation functions away from the crossing-symmetric point, which avoids numerical challenges associated with evaluating conformal blocks at different configurations The use of Hilbert space positivity provides an alternative route to imposing selection rules without explicitly decomposing the correlator into conformal multiplets This method successfully reproduces bounds comparable to those obtained from the conventional conformal block bootstrap at similar derivative orders The joint positivity constraint in higher dimensions leads to substantially improved bounds, constraining the correlator to within 1% of the Ising correlator estimate over a displayed range at Λ = 13 Pruning redundant reasoning tokens The construction is based on an elementary observation: any finite collection of states created by local operator insertions defines a positive semidefinite Gram matrix, which is a consequence of reflection positivity This leads to a hierarchy of semidefinite programs (SDPs) whose constraints can be systematically strengthened by enlarging the space of probe states The Gram matrix elements are computed by differentiating the overlap ⟨Ψ(c', s')Ψ(c, s)⟩ = G(u, v) − 1 with respect to c and s How the training signal is built The constraints are systematically strengthened by considering probe spaces constructed using “words” of conformal generators up to length L These probe spaces allow for the systematic computation of all entries of the Gram matrix, which are then converted into derivatives of the correlator with respect to the cross ratios via a change of variables Spectral assumptions can be incorporated straightforwardly by constructing a gap-localizing Gram matrix G gap = ⟨v i(D − ∆gap)v j⟩ ⪰ 0 Pruning redundant reasoning tokens The method is related in spirit to the quantum-mechanical bootstrap, where algebraic relations among operators are combined with Gram matrix positivity to constrain physical observables and spectra The familiar unitarity bounds on conformal primaries follow from the positivity of the multiplet Gram matrix, constructed using P† µ = K µ and the positivity of ⟨O a K µP ν Ob and ⟨O a K 2P squared Ob The constraints naturally form a nested hierarchy of semidefinite programs (SDPs) built from the probe spaces V0 ⊆ V1 ⊆ V2 ⊆ · · · Where the gains come from The antipodal configuration, in almost all cases, leads to stronger bounds on both the correlator and the gap, with most of them matching the bounds on the gap obtained from the conventional bootstrap within bisection tolerance The geometric interpretation shows that these generators span all possible infinitesimal variations of the two insertion points This observation suggests a recursive generalization of our construction across different kinematic configurations The method is capable of reproducing bounds comparable to those obtained using conformal blocks at similar derivative orders, even though it does not explicitly resolve conformal multiplets B SDP feasibility problem The gap maximization bounds presented in this work were obtained using bisection, with excluded gaps certified by dual feasibility The results show that the Gram matrix bootstrap can reproduce bounds comparable to those obtained from the conventional conformal block bootstrap at similar derivative orders The joint positivity constraint in higher dimensions leads to substantially improved bounds, constraining the correlator to within 1% of the Ising correlator estimate over a displayed range at Λ = 13 Discussion The framework suggests that much of the constraining power of the conventional conformal bootstrap can be accessed directly through these fundamental consistency conditions The flexibility of our framework also opens several directions for future development, particularly for bootstrap problems where the explicit computation of conformal blocks becomes challenging A natural generalization is to consider states created by non-identical or spinning local operators, leading to mixed-correlator and spinning-correlator bootstrap The method has already demonstrated that in d = 1, even a 2 × 2 Gram matrix can yield nontrivial bounds on the spectral gap The paper demonstrates that conformal symmetry, crossing symmetry, and unitarity can be directly utilized in the bootstrap by probing the CFT Hilbert space through Gram matrix positivity Computing Gram matrix elements The Gram matrix can be further simplified by exploiting the little group of the antipodal configuration, which describes the spatial transformations that preserve the insertion points, including their exchange permitted by Bose symmetry This reduction in d = 3 involves decomposing the Gram matrices into independent blocks labeled by h and η, with an additional reflection parity organization for h = 0 The resulting decomposition reduces the size of the matrices in the SDP without weakening them B SDP feasibility problem The Gram matrix can be further simplified by exploiting the little group of the antipodal configuration, which describes the spatial transformations that preserve the insertion points, including their exchange permitted by Bose symmetry This reduction in d = 3 involves decomposing the Gram matrices into independent blocks labeled by h and η, with an additional reflection parity organization for h = 0 The resulting decomposition reduces the size of the matrices in the SDP without weakening them Where the gains come from The method is capable of reproducing bounds comparable to those obtained using conformal blocks at similar derivative orders, even though it does not explicitly resolve conformal multiplets The joint positivity constraint in higher dimensions leads to substantially improved bounds, constraining the correlator to within 1% of the Ising correlator estimate over a displayed range at Λ = 13 The antipodal configuration improves convergence particularly at larger external weights, with most of them matching the bounds on the gap obtained from the conventional bootstrap within bisection tolerance The joint positivity constraint in higher dimensions leads to substantially improved bounds, constraining the correlator to within 1% of the Ising correlator estimate over a displayed range at Λ = 13 Pruning redundant reasoning tokens The Gram matrix can be further simplified by exploiting the little group of the antipodal configuration, which describes the spatial transformations that preserve the insertion points, including their exchange permitted by Bose symmetry This reduction in d = 3 involves decomposing the Gram matrices into independent blocks labeled by h and η, with an additional reflection parity organization for h = 0 The resulting decomposition reduces the size of the matrices in the SDP without weakening them Where the gains come from The antipodal configuration improves convergence particularly at larger external weights, with most of them matching the bounds on the gap obtained from the conventional bootstrap within bisection tolerance The geometric interpretation shows that these generators span all possible infinitesimal variations of the two insertion points This observation suggests a recursive generalization of our construction across different kinematic configurations The method is capable of reproducing bounds comparable to those obtained using conformal blocks at similar derivative orders, even though it does not explicitly resolve conformal multiplets Where the gains come from The method is capable of reproducing bounds comparable to those obtained using conformal blocks at similar derivative orders, even though it does not explicitly resolve conformal multiplets The joint positivity constraint in higher dimensions leads to substantially improved bounds, constraining the correlator to within 1% of the Ising correlator estimate over a displayed range at Λ = 13 The antipodal configuration improves convergence particularly at larger external weights, with most of them matching the bounds on the gap obtained from the conventional bootstrap within bisection tolerance Where the gains come from

Title and authors: Kai: So, focusing on what they summarize in "Conformal Bootstrap Unblocked," they are essentially arguing that you can derive non-perturbative constraints on a conformal field theory just by looking at the structure of states generated by local operators and applying symmetry rules.

Mira: They take the elementary idea that any set of states from local insertions forms a positive semidefinite Gram matrix because of reflection positivity, and then use conformal symmetry to express those matrix elements in terms of cross ratios, which are further constrained by crossing symmetry.

Lev: So they are using the structure of these states—these Gram matrices—as the primary object to constrain the physics, instead of relying on the operator product expansion to define how those states combine.

Kai: Right. They build a hierarchy of semidefinite programs where you can systematically strengthen your constraints by increasing the length of words used in those probe spaces, or by including translation generators to construct these probe spaces.

Mira: And they also mention using a gap-localizing Gram matrix to incorporate spectral assumptions straightforwardly, which means you can test for positivity based on whether there's a gap in the spectrum of that matrix.

Lev: That sounds like you're trying to build a bridge between the continuous constraints from symmetry and the discrete requirements of having a physical spectrum with an actual energy gap.

Kai: The paper demonstrates that by maximizing an auxiliary variable subject to Gram matrix positivity conditions, they can find bounds on the spectral gap, and they use dual feasibility to certify those excluded gaps.

Mira: And when looking at their results in two and three dimensions for the Ising model, they found that imposing Bose symmetry up to spin six is enough to resolve those kinks in the critical theory.

Lev: So if you can resolve a known physical feature like a kink using only these positivity constraints and basic symmetry assumptions, that's a strong result for checking consistency.

The paper's summary: Kai: Now let's talk about the specific improvements they highlight in this work. One of the main gains is constraining correlation functions away from the crossing-symmetric point by using joint Gram matrices from probe states associated with different kinematic configurations.

Mira: That means you get additional constraints at new kinematic points, which tightens up how much freedom those correlators have compared to when you only look at one configuration.

Lev: That's useful because it means the bounds aren't just based on one set of symmetries; they are based on the interplay between different geometric setups, which sounds more robust.

Kai: Another improvement is using an operator A to introduce a second gap assumption for subleading operators, where A separates scalar primaries from all other states, and this allows for constraints like the one involving D minus epsilon squared terms being non-negative.

Mira: That's interesting because it moves beyond just bounding the leading dimensions and starts constraining how the next level of operators behaves relative to those leading ones.

Lev: If you can constrain both the leading and subleading physics simultaneously using these positivity conditions, that’s a much more comprehensive way to check the theory.

Kai: They also show they can strengthen bounds on OPE coefficients by projecting out a specific operator and focusing the Gram matrix positivity constraint only on the remaining orthogonal spectrum.

Mira: So, instead of calculating everything about one operator at a time, you isolate one piece and use positivity to get a lower bound on other things, like that lambda squared sigma epsilon term.

Lev: That sounds like a way to get concrete bounds on coupling constants or interaction strengths without having to perform the full calculation of the OPE coefficients in every possible channel.

The paper's improvements: Kai: So wrapping up with "Conformal Bootstrap Unblocked," the main point is that you can use conformal symmetry, reflection positivity, and crossing symmetry directly to probe the Hilbert space through Gram matrix positivity without needing explicit conformal blocks.

Mira: Essentially, they show that these fundamental consistency conditions have enough power to reproduce bounds comparable to what we get from the conventional bootstrap at similar derivative orders, even without resolving all the conformal multiplets explicitly.

Lev: For me, I think the most practical aspect is how this method allows you to test those consistency conditions in a way that might be more scalable for real hardware constraints because it relies on these positivity checks rather than needing to fully compute and decompose the blocks first.

Kai: And they've shown that in three dimensions, just imposing Bose symmetry up to spin six is sufficient to resolve the kinks in the Ising model, which gives a concrete result for a known interacting CFT.

Mira: It opens up several directions for future work, like extending this approach to higher-point correlation functions where you can use the positivity of states created by multiple operator insertions as additional consistency conditions.

Lev: I think extending it to Lorentzian kinematics and looking at connections with moment problems is where the real challenge lies, but this framework gives us a solid foundation to start that exploration.

Conclusion: Kai: So, "Conformal Bootstrap Unblocked" is about using the structure of states—those Gram matrices—and basic symmetries like reflection positivity and crossing symmetry to get non-perturbative bounds on a conformal field theory.

Mira: Right, it’s about bypassing the explicit operator product expansion by looking at how states are built from local operators and imposing constraints through those state overlaps.

Lev: I mean, for hardware runs, the real gain is seeing how these positivity checks translate into actual constraints on the spectral gap of our system when we're dealing with finite resources.

Kai: They did this by building a hierarchy of semidefinite programs where you can strengthen your probe spaces just by increasing the length of words used to construct those states.

Mira: And they showed that in two and three dimensions, for things like the Ising model, imposing Bose symmetry up to spin six is enough to resolve those kinks we see in the theory.

Lev: If you're running this on real hardware, I'm thinking about how computationally heavy those Gram matrix calculations are when you start adding all that probe space complexity.

Kai: Well, they did manage it by exploiting the little group of the antipodal configuration to decompose those matrices into smaller blocks based on helicity and parity.

Mira: That decomposition actually helped reduce the size of those semidefinite programs, which is a big deal for feasibility testing.

Lev: Reducing matrix size is always good for error correction overhead, but I wonder if that decomposition always holds up when you move to higher dimensions where the constraints get tighter.

Kai: They also found that joint positivity in higher dimensions leads to substantially improved bounds on the correlator, sometimes constraining it within one percent of the Ising estimate over a certain range.

Mira: That’s a big step because it means you aren't just getting one kind of bound; you’re getting constraints from two different points in the theory at once.

Lev: If we can tighten those bounds, that helps us set better error thresholds for any quantum simulation we try to run on these lattice models.

Kai: So, the paper proves that conformal symmetry, crossing symmetry, and unitarity are direct tools for probing the Hilbert space through Gram matrix positivity.

Mira: It’s a nice way to get consistency checks without having to explicitly figure out all the conformal blocks first.

Lev: It sounds like a solid way to get non-perturbative information when we don't have the luxury of calculating every single diagram in the full OPE expansion.

Kai: That’s what "Conformal Bootstrap Unblocked" does, it shows us how to use these fundamental consistency conditions directly.

Mira: It's a powerful framework for constraining correlation functions away from the crossing-symmetric point.

Lev: I think we should keep an eye on this method as we look into ways to handle higher-point correlation functions, because that’s where it gets really interesting for our hardware constraints.

More episodes

← Home