research35 min studyLesson 800 in Physics

Bulk reconstruction and quantum error correction questions

Explain redundant logical reconstruction within a code subspace without claiming independent copies or unrestricted information access.

Before you begin

On this page

The question for today

What you’ll understand

State reconstruction as an operator relation on a selected code subspace.
Check a finite qutrit erasure code and a contrasting failed qubit encoding.
Separate conditional reconstruction from efficient decoding and empirical interior claims.

Bulk reconstruction asks how suitable bulk observables can be represented using boundary degrees of freedom in a holographic model. Quantum error correction explains how different boundary regions can represent the same logical operation on a restricted code subspace. This is redundancy of an encoding, not independent copies of an unknown quantum state. The interpretation requires a specified state regime, algebra and approximation order; it does not grant an observer unrestricted or efficient access to every black-hole interior or establish an empirical quantum-gravity experiment.

Restrict the operator identity to the code

An isometric encoding maps a chosen logical Hilbert space into a larger boundary Hilbert space. A reconstructed boundary operator need only agree with the logical operation after acting on encoded states. Outside that image its action can be different, so multiple boundary representations need not be equal as operators on the entire microscopic Hilbert space. The code subspace in semiclassical applications typically restricts excitations and backreaction around a specified class of backgrounds.

Subregion reconstruction is associated with the appropriate entanglement wedge under semiclassical and algebraic assumptions. A large enough change of state or energy can change the wedge or invalidate one chosen reconstruction map. Gravitational dressing and algebra centers also complicate naive tensor factorizations of bulk regions. Consequently a finite-dimensional factorized code is an instructive model, not a literal complete factorization of the Hilbert space of quantum gravity into independent spatial cells.

V^\dagger V=I_{\rm code},\qquad V\mathcal O_{\rm code}=\mathcal O_A V\quad\text{on the chosen code subspace}

V is an isometry and OA acts on an available boundary subsystem A. The equality is a restricted encoding relation, not equality of all microscopic operators or a guarantee that OA can be implemented efficiently.↔ Scroll wide equations horizontally.

Use a code that actually corrects erasure

A qutrit has three basis states, allowing a simple code that stores one logical qutrit in three physical qutrits. The displayed encoding adds basis labels modulo three. Any single physical qutrit has a maximally mixed reduced density matrix independent of the logical state, including the vanishing of logical off-diagonal partial traces. The complementary pair can therefore recover the logical information. Checking only the diagonal logical basis states would be insufficient to establish protection of arbitrary superpositions.

The key property is decoupling of the erased subsystem from logical information, including a possible external reference. A recovery operation on the complementary pair can then reconstruct the logical state in this exact finite code. This differs from a classical repetition intuition: a qubit encoding into three identical basis labels protects some bit-flip information but loses an unknown relative phase after an erasure. An encoding must be tested against the actual error model rather than judged only by apparent repetition.

|i_L\rangle=\frac1{\sqrt3}\sum_{r=0}^2|r,\ r+i,\ r+2i\rangle,\quad i=0,1,2\quad(\text{all labels mod }3),\qquad \operatorname{Tr}_{23}|i_L\rangle\langle j_L|=\delta_{ij}\frac{I_3}{3}

The displayed exact qutrit code corrects one known-location erasure. The off-diagonal trace condition is essential for preserving arbitrary logical superpositions and entanglement with a reference.↔ Scroll wide equations horizontally.

Distinguish redundancy, recovery cost and information claims

Different overlapping boundary pairs can implement a logical operation in the qutrit code, just as multiple boundary regions can have related reconstruction roles in a holographic code picture. This does not violate no-cloning because the purported recovery regions overlap and their representations agree only on the encoded subspace. One cannot treat them as independent tensor factors each holding a freely accessible copy of the unknown logical state.

Exact tensor-network codes capture selected geometric and algebraic features but do not automatically reproduce continuum dynamics, arbitrary gravitational states or all finite-N corrections. Existence of a recovery map also does not prove practical decoding efficiency or access to experimentally unavailable boundary observables. A pure-qubit lab can distinguish relative phase from basis probabilities; it cannot encode a qutrit, implement erasure recovery, reconstruct a bulk operator or settle the information-loss problem of a general evaporating black hole.

\rho_E\ \text{independent of logical input, including logical coherences}\quad\Longrightarrow\quad\text{erasure of }E\text{ is correctable on the exact finite code}

This concise criterion presumes the full decoupling condition, including a reference system; for general observable-algebra codes and approximate gravitational reconstruction, the statement needs corresponding algebraic and error qualifications.↔ Scroll wide equations horizontally.

Work it through

Check coherence in a three-qutrit code

Encode the logical state $(|0_L\rangle+|1_L\rangle)/\sqrt2$ using the displayed code. Compute the first physical qutrit's reduced density matrix and its entropy. Explain why a single qutrit cannot reveal the logical phase.

Each diagonal logical term has partial trace $I_3/3$ . In the cross term, matching both second and third qutrit labels would require simultaneously r = r′ + 1 and r = r′ + 2 modulo three, which is impossible.
The two cross partial traces therefore vanish. The reduced state is $\tfrac12(I_3/3+I_3/3)=I_3/3$ , with three eigenvalues one third.
Its entropy is $-\sum_{a=1}^3(1/3)\ln(1/3)=\ln3$ . Changing the relative logical phase leaves the single-qutrit state unchanged. The missing information is recoverable from a complementary pair in this exact code, not from the erased qutrit by itself.

The result

The first qutrit is I3/3 with entropy ln 3, independent of the logical relative phase.

The living notebook

Explore the model

Interactive lab

Interactive lab · quantum

Phase is not fixed by basis populations

Reset and set BOTH Mixing angle θ = 90° and Relative phase φ = 0°; predict equal basis probabilities and positive Bloch x at time zero. Reset and set θ = 90° and φ = 180°; predict the same probabilities and negative Bloch x. This pure qubit distinguishes phase from populations only; it is not the qutrit code, an erasure channel or a bulk reconstruction algorithm.

Bloch vector / P(0)P(1)Phase latitude

Probability |0⟩: 71.1%
Probability |1⟩: 28.9%
Bloch (x, y, z): 0.78, 0.45, 0.42

Mixing angle θ65°Relative phase φ30°

t = 0 s

A normalized two-level pure state, not an interacting field simulation. Playback adds one radian per second to the relative phase; basis probabilities stay fixed. Parameter changes pause playback and recompute the model at the current elapsed time. Reset restores both parameters and returns time to zero. Playback starts paused. Adjust either parameter, then choose Play or Step.

Look forPreserving an unknown quantum state requires its coherences, beyond matching basis probabilities.

Pause & reason

Check your understanding

Interactive checks need JavaScript. You can still read the explanation.

Read the checkpoint explanation

The logical equality is restricted to the chosen code. Redundant overlapping representations do not by themselves imply cloning, unrestricted state coverage or efficient decoding.

Your turn

Try it yourself

Consider the different qubit encoding $|0_L\rangle=|000\rangle$ , $|1_L\rangle=|111\rangle$ . Compare the states of qubits two and three after erasing qubit one from the logical plus and minus superpositions. Can this encoding recover an arbitrary logical phase from the remaining pair?

Take a moment to make a prediction or work it out on paper. Open a hint when you need one.

Give me a hint

Tracing out the first qubit kills cross terms because ⟨0|1⟩ = 0, even though the encoded full states were orthogonal.

Compare with the solution

Both logical signs give the same remaining state, $\tfrac12(|00\rangle\langle00|+|11\rangle\langle11|)$ . The original relative sign is absent from the retained pair, so no operation on that pair can distinguish and recover both unknown logical inputs. This qubit repetition encoding therefore fails the stated erasure task for arbitrary quantum superpositions, despite retaining classical basis-label information. It is not interchangeable with the three-qutrit erasure code or evidence that holographic reconstruction supplies independent cloned states.

Keep the idea

Take it with you

Bulk reconstruction is a restricted logical-operator statement.
Quantum erasure correction must preserve coherences and reference entanglement.
Toy-code recovery and semiclassical evidence do not establish unrestricted efficient information access.

Read further

Original sources and deeper explanations to follow at your own pace.

Optional progress, saved in this browser when available. Revisit any lesson at any time.

Lesson not marked complete.