Steven GellerQuantum Computing, End to End

Book contents

Type at least two characters.

    Current section

    Part I. Mental Model

    1. Computation Is Physical
    2. Classical Information, Gates, Circuits, and Reversibility
    3. Probability vs Amplitude
    4. Qubits Without Mysticism
    5. Measurement and What It Does Not Mean
    6. Interference as the Engine of Quantum Algorithms
    7. Entanglement Without Faster-Than-Light Myths
    All chaptersGlossaryLabsSourcesAbout

    Part I. Mental Model · Chapter 5

    Measurement and What It Does Not Mean

    Measurement is where a quantum computation hands classical data back to you — and where most popular explanations go wrong. This chapter pins down what the operation actually does, what it costs the state, and which three myths about it to retire on sight.

    Steven Geller
    In this chapter 7 sections

    Reader question. How does the choice of measurement determine both the outcome distribution and the state available afterward?

    A measurement asks a basis- or observable-specific question, samples an outcome by the Born rule, and conditions the post-measurement state on that outcome; repeated shots estimate that experiment's distribution but cannot repair a wrong basis, circuit, or noise model.

    Scope and non-goals.
    • Interpretive claims about collapse or consciousness are outside the operational model used here.
    • Generalized measurements, tomography, and continuous measurement are deferred.
    One state, two questions, two distributions One state, two questions, two distributions |+> measure in Z basis{ |0>, |1> } measure in X basis{ |+>, |-> } 0P = 1/2 1P = 1/2 +P = 1 two outcomes, equal odds one outcome, certain Same prepared state in both rows. Only the basis; the question; changed.
    Figure 5.1. Notice that nothing about the state changes between the two rows; only the basis does. A measurement answers the question you actually asked, not the one you meant to ask.

    A measurement is a specified question

    Define basis, projectors, outcomes, and repeated preparation.

    Measurement is a physical interaction that produces classical data, governed by two inputs: the quantum state and the measurement basis you chose. For a qubit state ∣ψ⟩=α∣0⟩+β∣1⟩\lvert\psi\rangle=\alpha\lvert0\rangle+\beta\lvert1\rangle measured in the computational basis, the outcome is 0 with probability ∣α∣2\lvert\alpha\rvert^2 and 1 with probability ∣β∣2\lvert\beta\rvert^2. That is the whole rule, and it is simple.

    The implications are where people stumble. Measurement does not print the state. It does not reveal the amplitudes. It returns one outcome from the one measurement you chose to perform, and the unchosen questions stay unanswered; there is no second read.

    Measurement is also not bookkeeping you can postpone. Quantum algorithms are designed around what will be measured and when; the measurement plan is part of the algorithm, the way a return type is part of a function signature. Chapter 3's interference machinery only matters because it shapes what this step produces.

    Evidence boundary. Projective measurement probabilities and post-measurement states follow from projectors and the Born rule. [John Watrous] [Michael A. Nielsen]

    One state interrogated two ways

    Calculate Z- and X-basis statistics for the same input.

    For a normalized qubit state, computational-basis measurement produces P(0)=∣α∣2P(0)=\lvert\alpha\rvert^2 and P(1)=∣β∣2P(1)=\lvert\beta\rvert^2. Afterward, the state available for any future computation is the one the outcome implies: measure 0 and the state is ∣0⟩\lvert0\rangle in that basis; measure 1 and it is ∣1⟩\lvert1\rangle. The pre-measurement amplitudes are gone; spent to produce the outcome.

    Change the basis and you change the question. The same prepared state can yield sharply different outcome distributions under different measurement setups, not because the state is ambiguous but because the questions are different. Choosing a basis is choosing an observable, and no single run answers for the observables you did not pick.

    Repeated shots turn probabilities into observed frequencies, and frequencies estimate probabilities with sampling error. More shots shrink that error. They do not repair a wrong circuit, a wrong basis, bad calibration, or a flawed model; they estimate the experiment you actually ran, faithfully and at length.

    Evidence boundary. Measurement statistics depend on the chosen basis or observable. [Michael A. Nielsen] [John Preskill]

    Conditioning changes the next experiment

    Trace post-measurement states and a second measurement.

    The deepest trap is the hidden-value story: measurement reveals what the qubit "really was" all along, as an ordinary bit we simply hadn't read. That story fails on the worked example. If the qubit were secretly a definite 0 or 1, the X-basis certainty would have no explanation; and neither would the interference effects from Chapter 3 that let gates convert phase into outcome. Basis-dependent statistics are not a puzzle for the computational model; they are a prediction of it.

    Evidence boundary. Repeated shots estimate an output distribution but do not remove systematic circuit, basis, or device error. [John Preskill] [National Academies of Sciences]

    Pi2=Pi,Pr⁡(i)=Tr⁡(Piρ)=∣⟨i∣ψ⟩∣2P_i^2=P_i,\quad \Pr(i)=\operatorname{Tr}(P_i\rho)=\lvert\langle i\vert\psi\rangle\rvert^2(5.1)

    Notation contract: Projectors P_i, outcome probability Tr(P_iρ) or |<i|ψ>|², normalized conditional state; basis stated before every result.

    Shot count buys precision, not validity

    Separate sampling error from bias and model error.

    Measurement is a constrained read operation on a stateful system; one where the read itself changes the available state. You cannot inspect arbitrary private state without consequence. You select an observable, run the experiment, and collect classical samples. Designing those three choices is engineering, not philosophy.

    This shapes debugging practice. A simulator exposes amplitudes; hardware exposes measurement records. Treat simulation output, analytical checks, and hardware shots as three different evidence types with different failure modes, not as interchangeable printouts. The bug in many quantum programs lives in the gap between what the simulator showed and what the chosen measurements could ever reveal.

    Design the measurement backward from the quantity

    Turn observable choice into an experimental design step.

    Basis-dependent measurement experiment

    Acceptance contract for Basis-dependent measurement experiment
    FieldReader-visible record
    FormatRunnable notebook producing sequential-measurement records and confidence intervals
    VerificationSeeded tests verify analytic distributions, conditional second-shot outcomes, and convergence without conflating precision and accuracy.
    AvailabilitySource-embedded acceptance record; no separate download is claimed
    {
  "artifact": "Basis-dependent measurement experiment",
  "format": "Runnable notebook producing sequential-measurement records and confidence intervals",
  "acceptance_test": "Seeded tests verify analytic distributions, conditional second-shot outcomes, and convergence without conflating precision and accuracy.",
  "publication_state": "source-embedded contract and worked fixture"
}

    Scope boundary

    • Interpretive claims about collapse or consciousness are outside the operational model used here.
    • Generalized measurements, tomography, and continuous measurement are deferred.

    Depth commitment. Two-basis calculation, conditional state trace, sampling experiment, and design rule.

    Practice problem

    For |+>, compute and simulate Z→Z, X→X, and Z→X sequential measurements, including conditional second outcomes.

    Deliverable
    Analytic probability tree and a seeded 10,000-shot comparison.
    Pass condition
    The reference notebook checks every conditional probability within a declared confidence band and explains discrepancies.

    Verification record

    Expected solution form. Probability-tree solution plus reproducible simulation report.

    Model answer. Z then Z on |+> produces first result 0 or 1 with probability one half and repeats that result with certainty. X then X produces + twice with certainty. Z then X produces a fair Z result and, conditioned on either branch, a fair X result.

    Model result and check. CI runs the seeded experiment and applies binomial tolerance bands to each branch.

    Acceptance test. The reference notebook checks every conditional probability within a declared confidence band and explains discrepancies.

    Provenance

    Sources and review

    1. John Watrous. The Theory of Quantum Information. Cambridge University Press / University of Waterloo. 2018textbook
    2. Michael A. Nielsen and Isaac L. Chuang. Quantum Computation and Quantum Information. Cambridge University Press. 2010textbook
    3. John Preskill. Lecture Notes for Physics 219: Quantum Computation. California Institute of Technology. 2018graduate lecture notes
    4. National Academies of Sciences, Engineering, and Medicine. Quantum Computing: Progress and Prospects. National Academies Press. 2019consensus study report

    The load-bearing claims in the chapter are mapped inline to this registered source set. A citation supports only the bounded claim beside it.

    External review: Open.

    How review worksRevision historyReport or inspect errata

    Dependency graph

    Prerequisites for this chapter

    1. Qubits Without MysticismPart I. Mental Model · Chapter 4
    2. Inner Products, Orthogonality, and ProjectionPart II. Mathematical Core · Chapter 10

    Continue the thread

    1. Entanglement Without Faster-Than-Light MythsPart I. Mental Model · Chapter 7
    2. Quantum Key DistributionPart IV. Protocols and Algorithms · Chapter 26
    3. Interference as the Engine of Quantum AlgorithmsPart I. Mental Model · Chapter 6

    Cite this chapter