Holographic Quantum Error Correction

Quantum error correction clarifies why a low-energy bulk algebra can have several boundary representations and survive erasure of a boundary subregion. The useful statement is conditional: a declared encoding, noise channel, code domain, and logical algebra satisfy an exact or approximate recovery criterion. QEC language does not itself derive the AdS/CFT dictionary, the wedge geometry, gravitational dressing, or a nonperturbative finite-$N$ code.

Required background. Operator-Algebra Quantum Error Correction supplies the exact algebraic criterion. The Bulk Reconstruction Problem supplies the gravitational problem.

Helpful background. Leading Semiclassical JLMS and Code-Subspace Claims gives the recovery input. Symmetry, Covariance, and QEC Constraints and Locality, Code Distance, and Causal Constraints state important obstructions.

Erasure as a holographic noise model

Let $V:\mathcal H_{\mathrm{code}}\to\mathcal H_A\otimes\mathcal H_{\bar A}$ be a regulated encoding and let erasure of $\bar A$ be

$$\mathcal E_A(\rho)=\operatorname{Tr}_{\bar A}(V\rho V^\dagger).$$

The complementary channel $\mathcal E_{\bar A}$ contains the information available to the erased region. For a logical algebra $\mathcal M$, exact correctability is equivalent to the complementary channel being insensitive to the noncentral degrees of freedom in $\mathcal M$. In the full-matrix case this reduces to

$$\mathcal E_{\bar A}(\rho)=\tau_{\bar A} \quad\text{for all }\rho\in\mathcal S_{\mathrm{code}},$$

which is the decoupling condition. If $A$ can reconstruct $\mathcal M$ and $\bar A$ can reconstruct $\mathcal M'$, the center $\mathcal M\cap\mathcal M'$ is the only information both may share exactly.

The original AdS/CFT formulation used this redundancy to resolve the apparent conflict between multiple boundary reconstructions: representatives differ as physical CFT operators but agree after projection to the low-energy code Almheiri, Dong, and Harlow 2015. Tensor-network codes demonstrate the mechanism exactly Pastawski et al. 2015; JLMS and recovery arguments provide the leading semiclassical gravitational evidence.

Reconstruction and perturbative bulk operators

Suppose a dressed scalar $\Phi_a$ lies in the wedge algebra of $A$. A perturbative smearing formula may construct a boundary representative $\Phi_A^{\mathrm{HKLL}}$ order by order. QEC adds a structural statement: if the erasure channel is correctable on the code, any two valid representatives obey

$$P\left(\Phi_A-\Phi_{A'}\right)P=0$$

in the exact model, or have a bounded projected difference in the approximate one. This is not equality in the full CFT algebra. It is also sensitive to dressing: changing the gravitational anchor can change the physical asymptotic charges and therefore the logical observable.

First application

Choose a code containing the vacuum and finitely many scalar wavepackets in $\operatorname{EW}(A)$. Model loss of $\bar A$ by $\mathcal E_A$. Compute the complementary outputs for a spanning set $\rho_i$ and define

$$\delta_{\bar A} =\sup_{i,j}\frac12 \left\lVert\mathcal E_{\bar A}(\rho_i)- \mathcal E_{\bar A}(\rho_j)\right\rVert_1.$$

Information–disturbance results turn small $\delta_{\bar A}$ into a recovery bound. Compare the recovered operator with the perturbative reconstruction by checking all matrix elements on the code, rather than matching one correlator. The calculation exhibits how a boundary erasure can leave a wedge algebra recoverable while the complementary region retains only center data.

Adversarial control

Use a noise channel unrelated to geometric erasure—for example, dephase a global charge—or enlarge the code until backreaction moves the wedge. The geometric QEC argument no longer supplies a recovery map. Likewise, impose an exact continuous symmetry on a finite-dimensional covariant code and demand exact local correction; symmetry constraints can forbid the desired code parameters.

The failure is informative: QEC does not restore a bulk interpretation once the encoding, error model, or code domain has changed. One must recompute correctability.

Validity, limitations, and further applications

The gravitational use assumes large $N$, $G_N/L^{d-1}\sim N^{-2}$, low code energy, controlled $\alpha'/L^2$ and KK truncation, and a specified order in bulk loops. Exact equality belongs to finite toy codes. At finite $N$, errors depend on code dimension, energy constraint, region, and norm; nonperturbative terms can be invisible to every finite order in $1/N$.

The strongest conclusion is a powerful interpretation and a family of conditional recovery results, not proof that gravity is an exact code. Continue to Approximate Finite-N Recovery, Alpha-Bits, and Error Bounds for quantitative errors and to QEC Evidence, Current Disputes, and Status for the current finite-$N$ dispute.

The chapter overview compares these assumptions and limitations.

For the chapter-wide comparison of assumptions, counterevidence, falsifiers, and claim ceilings, see the claim-domain table.

References

• Almheiri, A., Dong, X., and Harlow, D. (2015), “Bulk Locality and Quantum Error Correction in AdS/CFT,” Journal of High Energy Physics 2015(04), 163. DOI; arXiv:1411.7041.
• Harlow, D. (2017), “The Ryu–Takayanagi Formula from Quantum Error Correction,” Communications in Mathematical Physics 354, 865–912. DOI; arXiv:1607.03901.
• Pastawski, F., Yoshida, B., Harlow, D., and Preskill, J. (2015), “Holographic Quantum Error-Correcting Codes: Toy Models for the Bulk/Boundary Correspondence,” Journal of High Energy Physics 2015(06), 149. DOI; arXiv:1503.06237.