Instantons and Conformal Holography
Sebastian de Haro, Anastasios C. Petkou
TL;DR
This paper builds and analyzes a tractable AdS4/CFT3 subsector in which a conformally coupled bulk φ^4 theory is holographically related to a 3D boundary theory with a φ^6 interaction. Through radial quantization and careful handling of instanton backgrounds, the authors establish an exact bulk–boundary correspondence for the scalar sector, identifying the boundary operator O with dimension 1 as the renormalized composite ⟨φ⟩ ≡ φ^2 and showing Φ_hol = -O in the free case, extended to the interacting case with mode-mixing due to broken dilatations. They further show that bulk one-particle states map to boundary two-particle states in a diagonal sector, effectively yielding a tensor-product structure between bulk and boundary Hilbert spaces. The work suggests a controlled realization of conformal holography, with implications for higher-spin holography, instanton physics, and potential cosmological interpretations via instanton decay and de Sitter-like fluctuations.
Abstract
We study a subsector of the AdS_4/CFT_3 correspondence where a class of solutions in the bulk and on the boundary can be explicitly compared. The bulk gravitational theory contains a conformally coupled scalar field with a Phi^4 potential, and is holographically related to a massless scalar with a Phi^6 interaction in three dimensions. We consider the scalar sector of the bulk theory and match bulk and boundary classical solutions of the equations of motion. Of particular interest is the matching of the bulk and the boundary instanton solutions which underlies the relationship between bulk and boundary vacua with broken conformal invariance. Using a form of radial quantization we show that quantum states in the bulk correspond to multiply-occupied single particle quantum states in the boundary theory. This allows us to explicitly identify the boundary composite operator which is dual to the bulk scalar, at the free theory level as well as in the instanton vacuum. We conclude with a discussion of possible implications of our results.
