Appendix E — Holography and Large-N Primer

Holography and Large-N Primer

See also. This appendix is only the lightweight holography prerequisite for the present course. Students wanting a full treatment should use a dedicated AdS/CFT course and its source map; the present repository does not assume that external curriculum.

This appendix collects the holographic-side prerequisites for Semester II. It is intended for a student who knows QFT and operator algebras but has limited prior exposure to AdS/CFT or large-$N$ field theory. The primer is schematic — for a full treatment, consult the references in §E.10.

E.1 What Is AdS/CFT?

Maldacena’s conjecture (1997; Maldacena, Adv. Theor. Math. Phys. 2 (1998) 231) proposes a nonperturbative duality between

• Boundary side: a conformal field theory (CFT) in $d$ spacetime dimensions;
• Bulk side: a theory of quantum gravity (typically string theory) on a $(d+1)$-dimensional asymptotically AdS spacetime.

The duality is expected to be exact in a complete formulation. In practice we usually control only a code subspace and a perturbative dictionary, and bulk gauge invariance makes the phrase “every local bulk operator” too naive. The course will therefore label each use of the dictionary by its regime rather than treating the conjecture as a proved operator-by-operator theorem.

Working examples:

• $\mathcal{N}=4$ super-Yang–Mills with gauge group $\mathrm{SU}(N)$ in 4D $\leftrightarrow$ Type IIB superstring theory on ${\mathrm{AdS}}_5\times S^5$.
• 2D CFTs with large central charge $\leftrightarrow$ 3D gravity (or string theory) on ${\mathrm{AdS}}_3$.

For the course, we use AdS/CFT as a framework without doing string-theory calculations; we only need the algebraic features at large $N$.

E.2 Large-$N$ Factorization

In a $\mathrm{U}(N)$ or $\mathrm{SU}(N)$ gauge theory, a typical single-trace operator is a normalized trace of a word in matrix-valued fields. The required power of $N$ depends on the theory and on whether the operator has been centered; we choose it so that its connected two-point function remains $O(1)$.

For centered, normalized single-trace operators, large-$N$ factorization says that connected higher-point functions are suppressed. In the standard adjoint-gauge-theory normalization one schematically has

$$⟨{\mathcal{O}}_1\cdots {\mathcal{O}}_n{⟩}_{\mathrm{c}}=O(N^{2-n}),$$

with convention-dependent variants. Consequently, leading correlators obey generalized-free-field Wick factorization. This is a statement about a distinguished large-$N$ operator sector, not about every CFT observable.

E.3 Type III${}_1$ at Large $N$

The von Neumann algebra generated by an appropriate large-$N$ generalized-free single-trace sector can be type III${}_1$ in the representations studied by Leutheusser—Liu and Witten. The precise algebra, region or time band, state, and limiting procedure matter. This emergent algebra shares the structural type of many continuum-QFT local algebras, but the two statements should not be conflated.

At finite $N$, the global irreducible algebra of a regulated or complete boundary theory may be represented as type I. A CFT Hilbert space is not finite-dimensional merely because the spatial manifold is compact, and continuum local algebras at finite $N$ can still be type III. The claimed type change concerns the emergent large-$N$ single-trace algebra in a specified representation, not every finite-$N$ local algebra.

Physical reading (Sem II Wk 1): in the Leutheusser—Liu/Witten setting, the emergent type III${}_1$ sector and its outer modular flow organize the sharp semiclassical exterior. Reading this as the algebraic imprint of a horizon is a holographic interpretation. Type III structure alone neither proves that a horizon exists nor reconstructs an interior.

This is central to the Sem II program: Witten 2022 / CPW / AAJ all work in the type III${}_1$ regime obtained at large $N$. The crossed-product construction (Block D) is then the natural framework.

E.4 Eternal AdS Black Hole

The maximally extended AdS-Schwarzschild geometry has:

• Two asymptotic AdS boundaries (left $L$, right $R$);
• A bifurcation surface at the center (the horizon);
• Two interior regions (future, past) behind the horizon.

The Penrose diagram is a diamond with two asymptotic edges. The right boundary supports a CFT${}_R$; the left boundary supports a CFT${}_L$. The two CFTs are causally disconnected (no observer in one boundary can send a signal to the other), but the bulk geometry connects them via the wormhole.

E.5 Maldacena’s TFD Identification

Holographic identification (Maldacena 2003). The two-sided eternal AdS—Schwarzschild black hole is represented on the boundary by two CFT copies in the thermofield-double state at the Hawking inverse temperature:

$$|\mathrm{BH}{⟩}_{\mathrm{gravity}}\leftrightarrow |{\mathrm{TFD}}_{{\beta }_H}{⟩}_{{\mathrm{CFT}}_R\otimes {\mathrm{CFT}}_L},$$

with ${\beta }_H=1/T_H$, $T_H$ the Hawking temperature of the BH.

This parallels the regulated Rindler-mode picture of the Minkowski vacuum, although the continuum wedge statement is algebraic rather than a literal tensor product. The connected wormhole is the bulk dual of this particular holographic state. Entanglement by itself does not imply a smooth or traversable geometry.

E.6 The ADM Hamiltonian as Modular Flow

In the asymptotic AdS geometry, there is a notion of ADM Hamiltonian $H_{\mathrm{ADM}}$ — the energy as measured at the asymptotic boundary. For the eternal BH dual to a TFD, the ADM Hamiltonian on the right boundary is $H_R$ (and on the left, $H_L$).

In the regulated TFD purification, the right-algebra modular operator is

$${\Delta }_R=e^{-{\beta }_H(H_R-H_L)}.$$

On a right observable, $H_L$ commutes and the modular automorphism is the ${\beta }_H$-rescaled $H_R$ flow. In the semiclassical eternal-black-hole code subspace this is identified with the bulk Killing flow fixing the bifurcation surface. Generic-state and generic-region modular flows need not be geometric.

E.7 Witten 2022 in One Sentence

Witten’s central observation (arXiv:2112.12828):

At strict $N=\infty$, Witten’s extended right algebra contains a central energy-fluctuation variable and is not a factor. Perturbative $1/N$ corrections make that variable act nontrivially, producing the modular crossed product of the underlying type III${}_1$ factor and hence a type II${}_{\infty }$ factor. In the paper’s semiclassical state family, its entropy is defined up to a state-independent constant and is related to black-hole generalized entropy.

This is the Sem II Block 1 result. The Block D crossed-product machinery + the holographic dictionary together produce it.

E.8 CPW vs. CLPW

Two distinct papers, often confused:

CPW (Chandrasekaran—Penington—Witten 2022, arXiv:2209.10454): Large $N$ algebras and generalized entropy. The paper constructs a type II${}_{\infty }$ algebra for single-trace physics in a microcanonical large-$N$ setting, without relying on the perturbative $1/N$ mechanism of Witten’s paper. For its semiclassical states, the algebra entropy matches the generalized entropy of the bifurcation surface up to the allowed constant.

CLPW (Chandrasekaran—Longo—Penington—Witten 2022, arXiv:2206.10780): An algebra of observables for de Sitter space. Operators in a de Sitter static patch are gravitationally dressed to an observer’s worldline. Imposing the observer constraint and the bounded-below observer Hamiltonian amounts to a spectral projection of the crossed product and produces a type II${}_1$ factor with a maximum-entropy state. A type II${}_1$ factor is generally infinite-dimensional; compactness of the horizon alone is not the explanation.

The author lists overlap, but the physics and the algebra type are different:

• CPW: type II${}_{\infty }$, infinite-dim Hilbert space, eternal BH.
• CLPW: type II${}_1$, finite normalized trace and a maximum-entropy state, de Sitter static patch with an observer.

The course’s Sem II Block 2 is CPW; CLPW appears as a contrasting aside in Block 5 Wk 14.

E.9 Ahmad–Jefferson (Sem II Block 4)

AAJ v2 (arXiv:2501.01487) studies how the modular and crossed-product description changes when a unitary acts on the combined type III${}_1$ factor and its commutant. The main steps are:

1. Review the crossed-product entropy framework (§2).
2. Derive the covariant modular structure and the change in trace weights under a general unitary perturbation (§3).
3. Apply the result to black-hole evaporation and to the time-dependent Gao—Jafferis—Wall double-trace deformation (§4).
4. Organize the entropy correction through quadratic order: five additional terms are linear in the perturbation and fifteen are quadratic. Thus there are twenty terms through quadratic order, not twenty quadratic terms.

The perturbation strength, not the Connes cocycle itself, is the expansion parameter. Connes cocycles remain part of the modular toolkit, but the course keeps AAJ’s unitary-state deformation distinct from the separate finite-model Araki perturbation ${\rho }_V\propto e^{-(K+V)}$.

E.10 Further Reading

For a complete AdS/CFT introduction:

• Aharony, Gubser, Maldacena, Ooguri, Oz, “Large N field theories, string theory and gravity,” Phys. Rep. 323 (2000) 183.
• McGreevy, “Holographic duality with a view toward many-body physics,” Adv. High Energy Phys. 2010, 723105.
• Harlow, “TASI Lectures on the Emergence of Bulk Physics in AdS/CFT,” arXiv:1802.01040.

For the algebraic-QFT side of holography:

• Liu, “Lectures on entanglement, von Neumann algebras, and emergence of spacetime,” arXiv:2510.07017 — this is Sem II Block 3 connective tissue. The recommended primary text.
• Witten, “Gravity and the crossed product,” arXiv:2112.12828 §2 — large-$N$ setup.
• CPW arXiv:2209.10454 §§2–3 — two-sided large-$N$ algebra.

For the holographic entropy formula:

• Ryu & Takayanagi, “Holographic derivation of entanglement entropy from AdS/CFT,” PRL 96 (2006) 181602.
• Lewkowycz & Maldacena, “Generalized gravitational entropy,” JHEP 08 (2013) 090.

The full Sem II program assumes familiarity with this conceptual framework but does not assume the student can do explicit AdS/CFT calculations. The Block D + Bisognano–Wichmann + Week 14 dressed-entropy toolkit, plus the Liu lectures as connective tissue, is sufficient to follow Witten/CPW/AAJ.

E.11 What’s Used Where in Semester II

Block	Setting	Key holographic input
1 — Witten 2022	one-sided BH at large $N$	type III${}_1$ at large $N$; ADM Hamiltonian as modular flow generator
2 — CPW	microcanonical large-$N$ black-hole sector	extrapolate dictionary; bifurcation-surface generalized entropy
3 — Liu lectures	structural overview	geometric modular flow in symmetric settings; proposed algebraic ER=EPR reading
4 — AAJ	unitary perturbation of a factor plus its commutant	changed modular data and trace weighting; GJW application
5 — MSY	bulk-side complement	ANEC, Shapiro time advance, geometric backreaction

E.12 Honest Scoping

The course does not teach holographic QFT from scratch. Students who want to understand the bulk-side physics of Witten/CPW/MSY in depth should consult AdS/CFT reviews. The course’s algebraic-side toolkit is the necessary preparation for reading these papers critically, not for deriving their bulk-side claims independently.

For the course’s stated goal — preparing students to read the recent crossed-product / dressed-entropy literature in algebraic QFT — Block D + this primer + Sem II Block 3 (Liu) is the minimum sufficient holographic background.

For a holography curriculum proper, use a dedicated AdS/CFT course. The references below are enough for the claims made in this dossier, but they are not a replacement for that course.