A regularization scheme for the AdS/CFT correspondence

Wolfgang Mück, K. S. ViswanathanJournal of High Energy Physics 1999 · DOI

TL;DR

This paper proposes a novel regularization scheme to address a long-standing challenge in the AdS/CFT correspondence: calculating the divergent "contact terms" in the boundary conformal field theory (CFT) two-point functions. By introducing a specific regularization procedure, the authors demonstrate a method to systematically compute these otherwise problematic contributions, offering a more complete picture of the correspondence.

What Problem Does This Paper Address?

The AdS/CFT correspondence is a powerful theoretical framework that relates a quantum gravity theory in Anti-de Sitter (AdS) space to a conformal field theory (CFT) living on its boundary. A key aspect of this correspondence is the formula $e^{-I_{AdS}[\phi]} = \langle \exp \int dx \, \phi_0(x) O(x) \rangle$, which suggests that correlation functions in the CFT can be computed from the on-shell action of a field in AdS space satisfying boundary conditions.

However, this formal relationship faces two significant hurdles. Firstly, defining a well-posed Dirichlet boundary value problem in AdS space is tricky, as generic fields don't always propagate to the boundary. While this has been addressed by considering boundaries inside AdS, a second, more persistent issue remains: the calculation of singular contributions to correlation functions at "coincidence points" (when the points in the correlation function are the same). Previous attempts to handle these divergences have relied on interpreting correlators as distributions or adding boundary counterterms. This paper tackles the latter problem by proposing a regularization scheme that directly calculates these divergent contact contributions before any counterterms are introduced, aiming for a more fundamental understanding of the correspondence.

Methods

The authors refine the AdS/CFT prescription by introducing a regularization scheme. They achieve this by modifying the bulk-boundary kernel $K(x,y)$ in AdS space with a regulating factor $e^{-\mu k^2}$, where $k$ is the momentum and $\mu$ is a small parameter. This factor is designed to make integrals well-defined for all values of the radial coordinate $x^0$ and spatial separation $x-y$.

The core of their method involves:

  1. Introducing a Regulating Factor: The kernel $K(x,y)$ is modified with $e^{-\mu k^2}$ to ensure convergence of integrals.
  2. Calculating the Boundary Derivative: The CFT two-point function is related to the boundary normal derivative of this regularized kernel.
  3. Expanding the Kernel: For non-coincident points, they expand the regularized kernel using known series representations of Bessel functions.
  4. Identifying Analytic vs. Non-Analytic Terms: They show that the analytic terms in the expansion do not contribute to the finite-distance two-point function, while the leading non-analytic term yields a finite result.
  5. Handling Coincident Points: For coincident points ($x=y$), they directly evaluate the regularized kernel and its derivative, leading to a contact term that depends on the regularization parameter $\gamma$.

The paper emphasizes that the limits $\mu \to 0$ and $\epsilon \to 0$ (where $\epsilon$ is related to the boundary value) must be taken simultaneously, with $\mu$ being of order $\epsilon^2$.

Datasets

Not discussed in the paper.

Key Insights and Ideas

  • Direct Calculation of Contact Terms: The most significant insight is the development of a method to directly calculate the divergent contact contributions to CFT correlators, rather than circumventing them with counterterms or distributional interpretations.
  • Regularization as a Tool: The paper demonstrates that a specific regularization scheme can naturally handle these divergences, providing a more fundamental approach to the AdS/CFT correspondence.
  • Distinction Between Analytic and Non-Analytic Contributions: The analysis reveals that for non-coincident points, only the non-analytic terms in the kernel expansion contribute to the finite-distance correlator, while analytic terms vanish in the limit.
  • Contact Term as a Local Counterterm: For coincident points, the regularization leads to a contact term that can be compensated by adding a local counterterm to the AdS action. This suggests a deep connection between regularization and the structure of the action.
  • Conformally Coupled Scalar Field as a Test Case: The paper highlights that the conformally coupled scalar field ($\alpha = 1/2$) provides an exact check of their procedure, reinforcing its validity.

Key Results

  • For non-coincident points, the paper derives a finite contribution to the two-point function of the form $\frac{2\alpha \epsilon^{2\alpha-1}}{|\mathbf{x}-\mathbf{y}|^{2\alpha}}$, where $\epsilon$ is a parameter related to the boundary value.
  • For coincident points, the regularization leads to a contact term proportional to $\frac{\gamma}{\epsilon^{d-2}}$, where $\gamma$ is a regularization-dependent parameter. This term can be eliminated by adding a local counterterm of the form $-\frac{\gamma}{2} \int d^d x \, \phi(x)^2$.

Limitations and Future Work

  • Eliminating the Need for Counterterms: While the paper shows how to compensate for contact terms with counterterms, it suggests that the regularization scheme itself might serve as a regulator for the CFT two-point function, potentially eliminating the need for explicit counterterms. This is presented as an interesting avenue for further study.
  • Generalization to Other Fields and Correlators: The presented scheme is demonstrated for a scalar field and two-point functions. Extending this regularization to other fields (fermions, gauge fields) and higher-point functions would be a natural next step.
  • Understanding the Role of $\gamma$: The regularization-dependent parameter $\gamma$ for coincident points warrants further investigation to fully understand its physical implications.
  • Connection to Renormalization Group Flow: The interplay between the regularization scheme and the renormalization group flow of the CFT could be a fruitful area for future research.

Summary updated

BibTeX
@article{muck1999regularization,
  author = {Wolfgang Mück and K. S. Viswanathan},
  title = {{A regularization scheme for the AdS/CFT correspondence}},
  journal = {Journal of High Energy Physics},
  year = {1999},
  doi = {10.1088/1126-6708/1999/07/018},
}
Figure