diff --git a/.github/workflows/hosted-math.yml b/.github/workflows/hosted-math.yml new file mode 100644 index 0000000..9fab5a2 --- /dev/null +++ b/.github/workflows/hosted-math.yml @@ -0,0 +1,64 @@ +name: Hosted mathematics + +on: + pull_request: + paths: + - '**.md' + - 'verification/hosted_math/**' + - '.github/workflows/hosted-math.yml' + +permissions: + contents: read + +concurrency: + group: hosted-math-${{ github.event.pull_request.number }} + cancel-in-progress: true + +jobs: + capture: + if: github.event.pull_request.head.repo.full_name == github.repository + runs-on: ubuntu-24.04 + timeout-minutes: 25 + env: + CAPTURE_REVISION: ${{ github.event.pull_request.head.sha }} + steps: + - name: Check out the exact proposed source + uses: actions/checkout@11d5960a326750d5838078e36cf38b85af677262 + with: + ref: ${{ env.CAPTURE_REVISION }} + persist-credentials: false + - name: Set up Node + uses: actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020 + with: + node-version: '22' + - name: Install the isolated capture runtime + run: | + npm install --prefix "$RUNNER_TEMP/math-runtime" --no-audit --no-fund playwright@1.56.1 mathjax-full@3.2.1 + node "$RUNNER_TEMP/math-runtime/node_modules/playwright/cli.js" install --with-deps chromium + - name: Probe a representative hosted page + run: | + export NODE_PATH="$RUNNER_TEMP/math-runtime/node_modules" + node verification/hosted_math/probe_hosted_math.cjs \ + --root "$GITHUB_WORKSPACE" \ + --repository "$GITHUB_REPOSITORY" \ + --revision "$CAPTURE_REVISION" \ + --page README.md \ + --output "$RUNNER_TEMP/hosted-probe" + - name: Compare all hosted Markdown at both viewports + run: | + export NODE_PATH="$RUNNER_TEMP/math-runtime/node_modules" + node verification/hosted_math/capture_hosted_math.cjs \ + --root "$GITHUB_WORKSPACE" \ + --repository "$GITHUB_REPOSITORY" \ + --revision "$CAPTURE_REVISION" \ + --output "$RUNNER_TEMP/hosted-math" + - name: Retain actual reports and captures + if: always() + uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 + with: + name: hosted-math-${{ env.CAPTURE_REVISION }} + path: | + ${{ runner.temp }}/hosted-probe + ${{ runner.temp }}/hosted-math + if-no-files-found: error + retention-days: 14 diff --git a/README.md b/README.md index 5978041..405d17e 100644 --- a/README.md +++ b/README.md @@ -59,15 +59,20 @@ Start with the [connected scientific route](science/README.md), or go directly t [llms.txt](llms.txt) maps theorem, story and reproduction requests to those files. +## Learn the argument + +The [learning route](learning/README.md) uses John Watrous's *The Theory of Quantum Information* (2018) as its educational anchor. A [section-by-section bridge](learning/WATROUS_BRIDGE.md) and [four-label worked example](learning/WORKED_EXAMPLE.md) lead into [30 intrinsic milestones](learning/MILESTONES.md): 12 on background and preliminaries, followed by 18 on the theorem, matching proofs, costs and interpretation. The route builds from local calculations to the complete scientific claim. + ## Reproduce the finite checks With Python 3.10 or later, no additional packages are needed: ~~~bash python3 verification/check_science.py +python3 verification/check_learning.py ~~~ -The suite independently checks phase ensembles, occupation blocks, Kraus admissibility and factorization, exact batch probabilities, classical posteriors, charged source-clock schedules and phase-cell diffusion. The [verification guide](verification/README.md) gives its finite scope. These checks support the derivations; they are not an asymptotic proof, hardware experiment or priority certificate. +The science suite independently checks phase ensembles, occupation blocks, Kraus admissibility and factorization, exact batch probabilities, classical posteriors, charged source-clock schedules and phase-cell diffusion. The learning checker checks the map's structure and navigation. The [verification guide](verification/README.md) gives their finite scope. These checks support the derivations; they are not an asymptotic proof, hardware experiment or priority certificate. ## Scope and costs diff --git a/learning/MILESTONES.md b/learning/MILESTONES.md new file mode 100644 index 0000000..02d5b11 --- /dev/null +++ b/learning/MILESTONES.md @@ -0,0 +1,293 @@ +# Thirty milestones for phase persistence in complement sampling + +This map follows one scientific question: **how long must an unknown phase pattern persist for collective quantum samples to help produce an element outside their own support?** Read it together with the [Watrous bridge](WATROUS_BRIDGE.md), the [worked example](WORKED_EXAMPLE.md), and the [scientific overview](../science/README.md). + +There are exactly **30 milestones**: **12 background / preliminaries** milestones and **18 narrative / innovation** milestones. The first twelve establish the distinctions needed to read this particular source experiment: classical support versus quantum phase, a hidden table versus independent noise, full-state comparison versus one output event, and sample count versus a source clock. The remaining eighteen connect those foundations to the complete lower bound, the inherited recovery instrument, the refresh schedule, robustness, and resource accounting. “Narrative / innovation” names a part of the learning route; it does not claim that every ingredient in that part is new. + +The educational anchor is John Watrous, *The Theory of Quantum Information* (2018), using the author's [available manuscript and chapter files](https://cs.uwaterloo.ca/~watrous/TQI/). The precise anchor pointers below were checked in the official chapters: + +- **W2-states:** section 2.1.2, pp. 61–66: density operators, convex mixtures, pure states and composite systems. +- **W2-channels:** sections 2.2.1–2.2.2, pp. 72–90: channels and their representations, including Kraus operators. +- **W2-readout:** section 2.3.1, pp. 100–104, and section 2.3.2, pp. 105–113: measurements; the subsection “Nondestructive measurements and instruments,” pp. 111–113, supplies the flagged-process language. +- **W3-distance:** section 3.1.1, beginning p. 125: state discrimination, Theorem 3.4 and Proposition 3.5. +- **W7-symmetry:** section 7.1.1, beginning p. 391: permutation-invariant vectors, Proposition 7.1 and the symmetric occupation/bag basis beginning p. 393. + +The chapter files are [Chapter 2](https://cs.uwaterloo.ca/~watrous/TQI/TQI.2.pdf), [Chapter 3](https://cs.uwaterloo.ca/~watrous/TQI/TQI.3.pdf), and [Chapter 7](https://cs.uwaterloo.ca/~watrous/TQI/TQI.7.pdf). The manuscript page numbers identify the inspected layout. The bridge explains how these foundations enter the project; Watrous does not state the phase-persistence frontier, the complement posterior bound, or the project's collision-recovery schedule. + +## Background / preliminaries — M01 to M12 + +### M01 — One label outside a hidden set + +- **Phase:** Background / preliminaries. +- **Central question:** What is actually returned by a complement sampler? +- **Understanding target:** Specify a universe of size $`N=2M`$, a hidden support of size $`M`$, one classical output label, and unconditional success at least $`2/3`$. Distinguish returning one valid label from identifying the support or reconstructing a state. +- **Dependencies and story role:** Entry milestone. It fixes the output contract that every later upper and lower bound must address. +- **Conceptual connection:** A relational task can be easier than learning the complete hidden object; output size alone does not determine its information cost. +- **Source pointers:** W2-readout, section 2.3.1; [model sections 1–3](../science/model.md); [related work: the task and calibrated endpoint](../references/related-work.md). + +### M02 — Equal probabilities, different amplitudes + +- **Phase:** Background / preliminaries. +- **Central question:** How can two pure specimens produce identical basis samples but behave differently under interference? +- **Understanding target:** Write a uniform-magnitude subset state with phases and verify that computational-basis measurement removes those phases from the label probabilities. +- **Dependencies and story role:** Uses M01. It introduces the information that is absent from the ordinary-sample interface and motivates keeping quantum specimens. +- **Conceptual connection:** Born probabilities are one measurement's statistics, whereas a pure state specifies the statistics of every permitted measurement. +- **Source pointers:** W2-states, section 2.1.2; W2-readout, section 2.3.1; [model sections 1–2](../science/model.md); [worked example](WORKED_EXAMPLE.md). + +### M03 — A prior is a density operator + +- **Phase:** Background / preliminaries. +- **Central question:** What state does a receiver without the hidden phase table receive? +- **Understanding target:** Form the density operator averaged over a specified table prior while keeping the support fixed. Explain why this average describes the receiver's accessible experiment without saying that each realized specimen was mixed. +- **Dependencies and story role:** Uses M02. It establishes the operational object compared by the converse. +- **Conceptual connection:** A description conditioned on a hidden parameter and the observer's averaged state answer different questions. +- **Source pointers:** W2-states, section 2.1.2, especially the mixture and pure-state formulas; [model section 7](../science/model.md); [phase-twirl converse section 1](../science/phase-twirl-converse.md). + +### M04 — Reuse and refresh are different sources + +- **Phase:** Background / preliminaries. +- **Central question:** Why does averaging after making several copies differ from independently averaging each copy? +- **Understanding target:** Compare one hidden table reused within a block with a fresh independent table per specimen. Identify the correlations retained by the first experiment even when their one-copy averages agree. +- **Dependencies and story role:** Uses M03. It isolates the source property whose cost the main theorem measures. +- **Conceptual connection:** Equal one-time marginals do not determine a joint experiment. A random global phase also differs from independent address phases. +- **Source pointers:** W2-states, section 2.1.2, composite and correlated systems; [model sections 2, 5 and 7](../science/model.md); [interpretation](../science/interpretation.md). + +### M05 — Every receiver is a channel and a readout + +- **Phase:** Background / preliminaries. +- **Central question:** How can one statement cover collective measurements, ancillas, feedback and quantum memory? +- **Understanding target:** Represent the receiver as a quantum process followed by a classical output register. Recognize that unrestricted processing does not create extra source-dependent side information. +- **Dependencies and story role:** Uses M03 and M04. It prevents the converse from being read as a bound only on the proposed algorithm. +- **Conceptual connection:** A source-access lower bound can permit powerful computation while limiting the information delivered by that source. +- **Source pointers:** W2-channels, sections 2.2.1–2.2.2; W2-readout, section 2.3.2; [model section 3](../science/model.md); [phase-twirl converse sections 4 and 6](../science/phase-twirl-converse.md). + +### M06 — A flag does not erase the rejected copies + +- **Phase:** Background / preliminaries. +- **Central question:** How does heralded internal recovery lead to an unconditional complement-sampling experiment? +- **Understanding target:** Separate success probability, conditional output state, eventual classical success and total acquisition cost. Count copies consumed on rejected trials and specify an output when every flag fails. +- **Dependencies and story role:** Uses M01 and M05. It supplies the accounting needed to assess a rare-event recovery instrument. +- **Conceptual connection:** A trace-nonincreasing successful map is one branch of a complete instrument; selecting that branch is not a free experimental resource. +- **Source pointers:** W2-readout, section 2.3.2, “Nondestructive measurements and instruments”; W2-channels, section 2.2.2; [model section 4](../science/model.md); [coherent recovery: hard schedule](../science/coherent-recovery.md). + +### M07 — From a full-state bound to one output event + +- **Phase:** Background / preliminaries. +- **Central question:** What does trace distance say about the final probability of returning an outside label? +- **Understanding target:** Use half trace distance to bound the change in any measurement event, including after a receiver channel. Keep the factor-of-two convention explicit. +- **Dependencies and story role:** Uses M03 and M05. It is the bridge from source-state simulation to the actual classical-output lower bound. +- **Conceptual connection:** The converse compares complete input states so that one bound survives every allowed downstream decision. +- **Source pointers:** W3-distance, section 3.1.1, Theorem 3.4 and Proposition 3.5; [phase-twirl converse sections 2, 4 and 6](../science/phase-twirl-converse.md); [Watrous bridge](WATROUS_BRIDGE.md). + +### M08 — Occupations without measuring the labels + +- **Phase:** Background / preliminaries. +- **Central question:** Why can identical copies be expressed by label multiplicities while retaining coherence? +- **Understanding target:** Define a normalized symmetric occupation state as the superposition of all ordered strings with one histogram. Distinguish this basis change from measuring a histogram or sorting and discarding its records. +- **Dependencies and story role:** Uses M02 and M05. It gives a common language for both the converse and recovery. +- **Conceptual connection:** Permutation symmetry reduces redundant descriptions while preserving the phase-bearing amplitudes required by the algorithm. +- **Source pointers:** W7-symmetry, section 7.1.1, Proposition 7.1 and the bag basis; [phase-twirl converse section 1](../science/phase-twirl-converse.md); [coherent recovery: finite-batch instrument](../science/coherent-recovery.md). + +### M09 — The birthday weight inside a quantum state + +- **Phase:** Background / preliminaries. +- **Central question:** Why do ordinary multinomial probabilities appear in a coherent-copy calculation? +- **Understanding target:** Derive the squared occupation coefficients and the bound on the probability of a $`q`$-fold repeated label. Identify which probability is being bounded and why it does not describe a measured collision label. +- **Dependencies and story role:** Uses M08. It introduces the scale that later controls both lost distinguishability and successful phase recovery. +- **Conceptual connection:** A classical occupancy law can quantify the weight of a coherent subspace without imposing a classical receiver. +- **Source pointers:** W7-symmetry, section 7.1.1, for normalized occupation vectors; [phase-twirl converse sections 1–2](../science/phase-twirl-converse.md); [coherent recovery: exact success probability](../science/coherent-recovery.md). The birthday estimate is supplied in these project proofs. + +### M10 — Averaging phases selects characters + +- **Phase:** Background / preliminaries. +- **Central question:** Which occupation matrix elements survive independent random root phases? +- **Understanding target:** Show that a $`q`$th-root phase average preserves an entry precisely when its two occupation vectors agree modulo $`q`$ at every address. Compare that condition with equality under continuous phases. +- **Dependencies and story role:** Uses M02 and M08. It identifies the exact structural difference the converse must bound. +- **Conceptual connection:** Group-character orthogonality turns an unknown phase prior into a block structure, rather than an arbitrary loss of all coherence. +- **Source pointers:** W2-states, section 2.1.2, for averaged density operators; W7-symmetry, section 7.1.1, for the chosen basis; [phase-twirl converse section 1](../science/phase-twirl-converse.md); [related work: random phases](../references/related-work.md). The cyclic-character calculation is given explicitly in the project. + +### M11 — What an ordinary sample rules out + +- **Phase:** Background / preliminaries. +- **Central question:** After seeing several labels in the support, how likely is an unseen label to lie outside it? +- **Understanding target:** Condition a uniform balanced-support prior on the observed distinct-label set. Obtain the posterior success bound and explain why arbitrary processing of that classical transcript cannot improve its information. +- **Dependencies and story role:** Uses M01 and M05. It fixes the matched classical baseline needed by the quantum converse. +- **Conceptual connection:** The sample transcript excludes known labels but leaves all unseen labels exchangeable. +- **Source pointers:** W2-readout, section 2.3.1, for classical outcome probabilities; [phase-twirl converse section 5](../science/phase-twirl-converse.md); [related work: BBW26 posterior argument](../references/related-work.md). The posterior calculation is a project-specific use of the established complement-sampling baseline. + +### M12 — One complete source experiment + +- **Phase:** Background / preliminaries. +- **Central question:** Which assumptions must remain fixed when comparing receivers? +- **Understanding target:** State the known support size, fixed phase order, exact known block length, copy-only access, hard acquisition cap and allowed cross-block memory. Explain why “at most $`B`$ copies per table” cannot support the upper bound. +- **Dependencies and story role:** Integrates M01 and M04–M11. It closes the preliminary framework before the scientific argument begins. +- **Conceptual connection:** A persistence promise concerns the source clock; it is distinct from a receiver batch limit or storage restriction. +- **Source pointers:** W2-states, section 2.1.2; W2-channels, section 2.2.1; [model sections 2–4 and 8](../science/model.md); [implementation: acquisition and classical costs](../science/implementation.md). + +## Narrative / innovation — M13 to M30 + +### M13 — The calibrated one-copy endpoint + +- **Phase:** Narrative / innovation. +- **Central question:** How does one positive quantum sample become a complement sample? +- **Understanding target:** Apply the known uniform-state reflection and verify that balanced cardinality maps the positive support state exactly to the positive complement state. +- **Dependencies and story role:** Uses M02, M05 and M12. It establishes the inherited capability that unknown phases obstruct. +- **Conceptual connection:** An output absent from the measured input support can arise through interference; this is the original complement-sampling mechanism. +- **Source pointers:** W2-states, section 2.1.2; W2-readout, section 2.3.1; [model section 5](../science/model.md); [coherent recovery: subset to complement](../science/coherent-recovery.md); [related work: BBW26](../references/related-work.md). + +### M14 — Continuous phases give an exact sample simulator + +- **Phase:** Narrative / innovation. +- **Central question:** What can ordinary samples reproduce when a continuous phase table is hidden and frozen? +- **Understanding target:** Draw ordinary labels, prepare their normalized symmetric histogram state, and recover the complete continuous-phase averaged block exactly, including repeated labels. +- **Dependencies and story role:** Uses M08, M10 and M12. It supplies the inherited reference experiment for the new finite-alphabet converse. +- **Conceptual connection:** Quantum processing of classical data can preserve permutation coherence without receiving additional hidden-support information. +- **Source pointers:** W7-symmetry, section 7.1.1; W2-states, section 2.1.2; [phase-twirl converse section 3](../science/phase-twirl-converse.md); [related work: GMR25](../references/related-work.md). + +### M15 — A good occupation has no modular partner + +- **Phase:** Narrative / innovation. +- **Central question:** Why does agreement modulo $`q`$ force exact equality when every occupation is below $`q`$? +- **Understanding target:** Reconstruct the nonnegative-shift proof using equal total occupation. Verify both identical good-good blocks and the absence of good-bad cross terms. +- **Dependencies and story role:** Uses M09, M10 and M14. This is the load-bearing inference that permits a full-state comparison rather than a heuristic birthday argument. +- **Conceptual connection:** An arithmetic uniqueness statement identifies where finite phase alphabets first become distinguishable from continuous phases. +- **Source pointers:** W7-symmetry, section 7.1.1, for the occupation basis; [phase-twirl converse section 2, Lemma 1](../science/phase-twirl-converse.md); [related work: occupation-sector antecedents](../references/related-work.md). + +### M16 — Common positive blocks bound distinguishability + +- **Phase:** Narrative / innovation. +- **Central question:** How does a common good block yield a trace-distance bound with the correct constant? +- **Understanding target:** Write each state as the shared positive good component plus a positive residual with equal trace. Bound half trace distance by the bad occupation probability. +- **Dependencies and story role:** Uses M07 and M15. It turns the modular argument into a quantitative bound valid before every receiver. +- **Conceptual connection:** Positivity controls the entire state without conditioning away unsuccessful events or assuming a classical algorithm. +- **Source pointers:** W3-distance, section 3.1.1; W2-states, section 2.1.2; [phase-twirl converse section 2, equations (3)–(5)](../science/phase-twirl-converse.md). + +### M17 — Independent source blocks, unrestricted receiver memory + +- **Phase:** Narrative / innovation. +- **Central question:** How does the comparison survive a receiver that coherently combines all refresh blocks? +- **Understanding target:** Tensor the conditional-on-support block states, replace factors one at a time, and sum the trace-distance errors. Identify why independence is used for the source prior, not imposed on the receiver. +- **Dependencies and story role:** Uses M04 and M16. It introduces persistence length while retaining the full allowed receiver class. +- **Conceptual connection:** A hybrid comparison of joint input states can remain valid under downstream correlations created by quantum memory. +- **Source pointers:** W2-states, section 2.1.2, composite systems; W3-distance, section 3.1.1; [phase-twirl converse section 4](../science/phase-twirl-converse.md); [model section 3](../science/model.md). + +### M18 — Bound the actual complement label + +- **Phase:** Narrative / innovation. +- **Central question:** How do the posterior baseline and phase-state comparison constrain the receiver's final label? +- **Understanding target:** Combine the ordinary-sample success bound with the sum of block collision weights. Include adaptive early stopping by padding to the hard acquisition cap. +- **Dependencies and story role:** Uses M11 and M17. It proves a lower bound for the agreed classical output, rather than borrowing an incompatible coherent-state transformation bound. +- **Conceptual connection:** A weaker output contract requires its own payoff argument even when the source-state machinery is familiar. +- **Source pointers:** W3-distance, section 3.1.1, Proposition 3.5; W2-readout, section 2.3.2; [phase-twirl converse sections 5–6, Theorem 2](../science/phase-twirl-converse.md). + +### M19 — Two obstructions and one lower frontier + +- **Phase:** Narrative / innovation. +- **Central question:** Why must a successful receiver pay both a global collision scale and a refresh-limited scale? +- **Understanding target:** Bound the block sum once by total copies to power $`q`$ and once by total copies times $`B^{q-1}`$. Use a fixed success gap to derive the maximum of the two scales, capped by ordinary linear cost. +- **Dependencies and story role:** Uses M18. It establishes the asymptotic lower half of the principal theorem with constants uniform in $`B`$ for fixed $`q`$. +- **Conceptual connection:** An interaction between total resources and source persistence produces a frontier, not one universal birthday exponent. +- **Source pointers:** W3-distance, section 3.1.1, for the event-comparison foundation; [phase-twirl converse section 7](../science/phase-twirl-converse.md); [persistence frontier sections 1–3](../science/persistence-frontier.md). + +### M20 — An admissible inherited recovery instrument + +- **Phase:** Narrative / innovation. +- **Central question:** Why do the proposed successful Kraus operators define a physical process? +- **Understanding target:** Identify accepted occupations, the unique distinguished repeated label, the residual vector and the bounded falling-factorial coefficient. Verify that the sum of successful effects is at most the identity. +- **Dependencies and story role:** Uses M05, M06 and M08. It begins the constructive half with the actual inherited instrument, rather than an unspecified nonlinear operation. +- **Conceptual connection:** A nonlinear normalized target can arise on a heralded branch of a linear completely positive map. +- **Source pointers:** W2-channels, section 2.2.2; W2-readout, section 2.3.2, instruments; [coherent recovery: finite-batch instrument](../science/coherent-recovery.md); [related work: QCC26](../references/related-work.md). + +### M21 — Factorization removes the hidden phase + +- **Phase:** Narrative / innovation. +- **Central question:** Why may the successful residual registers be discarded without losing the recovered subset coherence? +- **Understanding target:** Cancel the occupation multiplicity factors and factor the successful vector into a coordinatewise powered output and one residual vector. Show that the $`q`$th power removes root phases. +- **Dependencies and story role:** Uses M20. It justifies coherent phase recovery, including the cleanup requirement that a measured collision label would violate. +- **Conceptual connection:** Residual independence, rather than a convenient notation for postselection, licenses the partial trace. +- **Source pointers:** W2-states, section 2.1.2, composite systems; W2-channels, section 2.2.2; [coherent recovery: factorization calculation](../science/coherent-recovery.md); [implementation: cleanup](../science/implementation.md). + +### M22 — The exact finite-batch success probability + +- **Phase:** Narrative / innovation. +- **Central question:** What does one complete recovery attempt cost and how often does it succeed? +- **Understanding target:** Sum the residual multinomial weights to obtain the falling-factorial success formula times a no-$`q`$-collision occupancy probability. Check that the answer is independent of the hidden phases and support labels. +- **Dependencies and story role:** Uses M09 and M21. It supplies the finite probability required to choose a charged schedule. +- **Conceptual connection:** The same ordinary occupancy law quantifies a useful coherent subspace in the upper bound and an inaccessible one in the converse. +- **Source pointers:** W7-symmetry, section 7.1.1, for normalized occupations; W2-readout, section 2.3.2, for successful trace; [coherent recovery: exact success probability](../science/coherent-recovery.md). + +### M23 — Choose a batch before it becomes too large + +- **Phase:** Narrative / innovation. +- **Central question:** Which batch sizes retain a useful success lower bound within one source phase block? +- **Understanding target:** Use the no-collision union bound to cap the residual batch near the frozen collision scale, then intersect that cap with the exact available block length. Recognize that oversized batches need not improve success. +- **Dependencies and story role:** Uses M22. It converts the exact instrument into a known, implementable choice of batch size. +- **Conceptual connection:** A collective advantage can depend on a bounded useful window, rather than monotonic improvement with more simultaneous inputs. +- **Source pointers:** W7-symmetry, section 7.1.1; [coherent recovery: exact probability and hard schedule](../science/coherent-recovery.md); [persistence frontier section 4](../science/persistence-frontier.md). + +### M24 — Count every phase-block tail + +- **Phase:** Narrative / innovation. +- **Central question:** How can repeated recovery attempts respect boundaries without an uncharged refresh or a hidden additive block cost? +- **Understanding target:** Pack complete attempts into each known block, charge discarded remainders, stop before an unused final tail, and derive the deterministic total-copy cap. Give a uniform fallback label if all attempts fail. +- **Dependencies and story role:** Uses M06, M12 and M23. It establishes a hard-budget upper bound for the actual source clock. +- **Conceptual connection:** A persistence resource must be acquired through the same interface that defines the lower bound. +- **Source pointers:** W2-readout, section 2.3.2, instruments and recorded outcomes; [coherent recovery: hard schedule](../science/coherent-recovery.md); [implementation: acquisition and classical costs](../science/implementation.md). + +### M25 — The classical branch closes the costly regimes + +- **Phase:** Narrative / innovation. +- **Central question:** Why is an ordinary-sample fallback part of the optimal quantum algorithm? +- **Understanding target:** Measure a known number of specimens, count unseen support labels, and use convexity to prove a constant complement success. Compare deterministic caps before choosing the coherent or classical branch. +- **Dependencies and story role:** Uses M11 and M12. It supplies the outer minimum and covers small blocks and finite dimensions. +- **Conceptual connection:** Matching a frontier can require different algorithms in different regimes; a quantum receiver is allowed to use ordinary sampling. +- **Source pointers:** W2-readout, section 2.3.1; [coherent recovery: classical fallback](../science/coherent-recovery.md); [persistence frontier sections 2 and 4](../science/persistence-frontier.md). + +### M26 — Read the complete persistence law + +- **Phase:** Narrative / innovation. +- **Central question:** What changes when the phase order or persistence scale changes? +- **Understanding target:** Combine the lower and upper bounds and locate the ordinary, refresh-limited and frozen regimes. Derive the sign and three-phase examples while keeping phase order fixed. +- **Dependencies and story role:** Uses M19, M24 and M25. It assembles the central result from its independently justified halves. +- **Conceptual connection:** Identical one-copy averaged density operators can support distinct optimal multicopy capabilities because their hidden-source correlations differ. +- **Source pointers:** W2-states, section 2.1.2, for the marginal/joint distinction; [persistence frontier sections 1, 5 and 6](../science/persistence-frontier.md); [interpretation](../science/interpretation.md). + +### M27 — Phase cells and an explicit success margin + +- **Phase:** Narrative / innovation. +- **Central question:** Does the recovery advantage survive bounded departures from exact roots? +- **Understanding target:** Assume $`|\delta_x|\le\eta/q`$, $`0\le\eta\lt \pi/2`$ and $`\cos^2\eta>2/3`$. Bound the mean residual phase after powering by the real part of the arc $`[-\eta,\eta]`$. Relate the resulting complement success to that fixed positive margin, and retain exact roots as the lower-bound subfamily. +- **Dependencies and story role:** Uses M13, M21, M24 and M26. It shows which robustness statement the theorem actually supports. +- **Conceptual connection:** A fixed phase promise can preserve a resource order without implying tolerance to arbitrary source noise or within-block drift. +- **Source pointers:** W2-states, section 2.1.2, pure-state amplitudes; W2-readout, section 2.3.1; [model section 6](../science/model.md); [coherent recovery: fixed-width phase cells](../science/coherent-recovery.md). + +### M28 — Rare success requires finer synthesis + +- **Phase:** Narrative / innovation. +- **Central question:** Why is a constant circuit approximation error insufficient for a rare recovery branch? +- **Understanding target:** Compare subnormalized successful outputs before conditioning, scale the batch error with its success lower bound, and control the full adaptive schedule with a hybrid over all attempted batches. +- **Dependencies and story role:** Uses M07, M24 and M27. It preserves the unconditional output guarantee when ideal rotations are replaced by finite gates. +- **Conceptual connection:** Normalizing a small successful state can magnify error; a complete-run bound also covers changed failure flags. +- **Source pointers:** W2-channels, section 2.2.2; W2-readout, section 2.3.2; W3-distance, section 3.1.1; [implementation: finite universal gates and complete hard schedule](../science/implementation.md). The circuit synthesis result is separately attributed there. + +### M29 — Price the source and the processor separately + +- **Phase:** Narrative / innovation. +- **Central question:** Which resource advantage has actually been established? +- **Understanding target:** Account for acquisition, batch width, known gates, classical schedules and source preparation. Explain the stated near-linear processing overhead and why it does not supply a free preparation circuit or a hardware crossover. +- **Dependencies and story role:** Uses M12, M24, M25 and M28. It fixes the implementation scope of the algorithmic conclusion. +- **Conceptual connection:** Supplied-copy complexity, coherent width, gate count and the cost of creating each specimen are distinct quantities. +- **Source pointers:** W2-channels, section 2.2.1, for the declared input/output interface; [implementation](../science/implementation.md); [model section 8](../science/model.md); [interpretation](../science/interpretation.md). + +### M30 — One connected scientific account + +- **Phase:** Narrative / innovation. +- **Central question:** What remains after the inherited ingredients, exact assumptions and full charges are stated together? +- **Understanding target:** Reconstruct the phase-persistence frontier from the source promise through the classical-output converse and attaining schedule. Identify the contributions of BBW26, GMR25, phase-moment work and QCC26, and keep unsupported extensions outside the conclusion. +- **Dependencies and story role:** Integrates M26–M29 and the proof routes they depend on. It closes the account at its established scope rather than adding an optional new research target. +- **Conceptual connection:** A precise operational connection can be the result while its foundational methods remain credited antecedents. +- **Source pointers:** W2-states, W2-channels, W2-readout, W3-distance and W7-symmetry as mapped in the [Watrous bridge](WATROUS_BRIDGE.md); [scientific overview](../science/README.md); [persistence frontier section 8](../science/persistence-frontier.md); [related work and attribution](../references/related-work.md). + +## Where to return for a specific gap + +Use the central question of a milestone to locate the missing part of the argument. M01–M12 resolve the experiment and language; M13–M19 reconstruct the converse; M20–M25 reconstruct the attaining algorithm; M26–M30 reconcile the frontier, robustness, implementation and attribution. The [worked example](WORKED_EXAMPLE.md) provides finite instances, and the [verification guide](../verification/README.md) identifies the actual checks and their limits. diff --git a/learning/README.md b/learning/README.md new file mode 100644 index 0000000..59f5f8a --- /dev/null +++ b/learning/README.md @@ -0,0 +1,22 @@ +# Learning route + +The educational anchor is **John Watrous, *The Theory of Quantum Information* (2018)**. Its [official author page](https://cs.uwaterloo.ca/~watrous/TQI/) identifies the edition and provides the author's manuscript. The textbook supplies the language of states, channels, measurements, distinguishability and symmetric tensors; the persistence theorem is established in this repository. + +Start with the [Watrous bridge](WATROUS_BRIDGE.md), which translates the selected textbook sections into the source model and supplies the extra occupation and phase calculations. Then follow the [30 milestones](MILESTONES.md) in order. The [four-label worked example](WORKED_EXAMPLE.md) makes the distinction between a one-copy average and repeated-source correlations explicit before returning to the asymptotic theorem. + +## Why 12 background milestones and 18 narrative milestones? + +M01–M12 build the complete language needed to ask the question: the support task, quantum versus classical data, correlated ensembles, instruments, unconditional success, trace-distance bounds, occupations, phase characters, posterior sampling and the source clock. M13–M30 then follow the actual argument from its simple endpoints through the unrestricted converse, the attaining instrument and schedule, and the implementation and interpretation limits. The longer second phase reflects the two matching proofs and their resource accounting. + +Within each phase, the route goes from a local calculation to its global role. The dependency list on each milestone names earlier steps that it uses. The map describes the project rather than a particular reader's progress. + +| Route | Use | +| --- | --- | +| [Watrous bridge](WATROUS_BRIDGE.md) | Selected sections, notation translation, and proof steps beyond the textbook | +| [Four-label example](WORKED_EXAMPLE.md) | A small source, its output, its phase correlations and its charged acquisition cost | +| [M01–M12](MILESTONES.md#m01--one-label-outside-a-hidden-set) | Background and preliminaries | +| [M13–M30](MILESTONES.md) | Narrative, proof and scientific contribution | +| [Scientific route](../science/README.md) | Full definitions and derivations | +| [Verification guide](../verification/README.md) | Reproducible finite and structural checks, with their limits | + +The result assumes a balanced support, fixed phase order, known blocks of exactly the stated number of consecutive source calls, and unconditional classical-label success. It charges failed batches and discarded tails. The phase-cell extension has an explicit arc and success margin. Receiver costs are separate from source preparation; the supplied-copy result does not establish an end-to-end classical-data runtime advantage. diff --git a/learning/WATROUS_BRIDGE.md b/learning/WATROUS_BRIDGE.md new file mode 100644 index 0000000..7b206f7 --- /dev/null +++ b/learning/WATROUS_BRIDGE.md @@ -0,0 +1,142 @@ +# From Watrous to phase persistence + +The selected educational anchor is **John Watrous, *The Theory of Quantum Information* (2018)**, published by Cambridge University Press in April 2018. The [author's official book page](https://cs.uwaterloo.ca/~watrous/TQI/) supplies a [pre-publication manuscript](https://cs.uwaterloo.ca/~watrous/TQI/TQI.pdf) for personal use, with a restriction against selling or redistributing it. This repository links to that manuscript; it does not reproduce or distribute it. + +The book provides the mathematical language for reading this project. The phase-persistence theorem is a research result developed in the [science route](../science/README.md), with its research dependencies identified in [related work](../references/related-work.md). Those papers and constructions are dependencies, not additional educational anchors. + +## A focused route through the selected book + +Section numbers and page numbers below refer to the author's standard-layout manuscript linked from the official page. + +| Official chapter | Verified sections to use | Connection to this project | +| --- | --- | --- | +| [Chapter 2](https://cs.uwaterloo.ca/~watrous/TQI/TQI.2.pdf) | §2.1.2, pp. 61–66; §§2.2.1–2.2.2; §§2.3.1–2.3.2, especially the instruments discussion on pp. 111–113 | Density operators and mixtures describe hidden phases; channels and measurements describe the receiver; an instrument records success while retaining a quantum output. | +| [Chapter 3](https://cs.uwaterloo.ca/~watrous/TQI/TQI.3.pdf) | §3.1.1, especially Theorem 3.4 and Proposition 3.5, pp. 128–130 | Trace norm controls changes in observable output probabilities. | +| [Chapter 7](https://cs.uwaterloo.ca/~watrous/TQI/TQI.7.pdf) | §7.1.1, especially Proposition 7.1 and the bag-indexed basis, pp. 391–394 | Identical pure copies admit an occupation basis, which makes phase averaging explicit. | + +## Translate the notation before the argument + +Watrous uses ordinary vectors and their adjoints rather than Dirac notation. The translation is: + +| Book notation | Notation used here | +| --- | --- | +| A classical alphabet $`\Sigma`$ and space $`\mathcal X=\mathbb C^\Sigma`$ | The known labels $`[N]=\{0,\ldots,N-1\}`$ and their $`N`$-dimensional register | +| Standard vector $`e_x`$; matrix unit $`E_{x,y}=e_xe_y^*`$ | $`\lvert x\rangle`$; $`\lvert x\rangle\langle y\rvert`$ | +| Pure state $`uu^*`$ | $`\lvert\psi\rangle\langle\psi\rvert`$ | +| Identity $`1_{\mathcal X}`$ and adjoint $`A^*`$ | $`I`$ and $`A^\dagger`$ | +| Measurement effect $`\mu(y)`$ | An effect $`E_y`$, with outcome probability $`\mathrm{Tr}(E_y\rho)`$ | +| $`n`$ tensor factors in Chapter 7; a bag $`\varphi\in\mathrm{Bag}(n,\Sigma)`$ | $`t`$ copies in one block; an occupation vector with entries $`n_x=\varphi(x)`$ | + +The last row prevents a useful ambiguity: the book's tensor count $`n`$ is not the universe size $`N`$. In this project $`N=2M`$, the hidden support $`S`$ has size $`M`$, $`q\ge2`$ is a fixed phase order, $`B`$ is the known number of consecutive source calls sharing one table, and $`T`$ is the hard total acquisition budget. An individual block contributes $`t\le B`$ observed copies. The receiver knows these parameters and the basis, but does not know $`S`$ or the phase table. + +## Mixture and product are different operations + +For a fixed hidden support and table, define + +```math +|\psi_{S,z}\rangle +=\frac1{\sqrt M}\sum_{x\in S}\omega_q^{z_x}|x\rangle, +\qquad \omega_q=e^{2\pi i/q}. +``` + +Averaging independent uniform table entries gives the one-copy state + +```math +\rho_S +=\mathbb E_z|\psi_{S,z}\rangle\langle\psi_{S,z}| +=\frac1M\sum_{x\in S}|x\rangle\langle x|. +``` + +The convex-mixture and tensor-product rules from Chapter 2 must be applied in the order specified by the source. A table drawn once for $`t`$ emissions gives + +```math +\Omega_{q,S}^{(t)} +=\mathbb E_z\!\left[ + (|\psi_{S,z}\rangle\langle\psi_{S,z}|)^{\otimes t} +\right], +``` + +whereas independent redraws at every emission give $`\rho_S^{\otimes t}`$. Their one-register reductions agree; their joint states need not. This is a correlation created by reusing a hidden classical parameter. No entangling source operation is assumed: conditioned on the table, the emitted copies are a product state. + +The [four-label worked example](WORKED_EXAMPLE.md) displays an off-diagonal entry that survives the first average and disappears in the second. + +## From the symmetric basis to the phase-specific step + +A vector $`|\psi\rangle^{\otimes t}`$ is unchanged by permutations of the copies. The bag basis in §7.1.1 becomes the normalized occupation state $`|n\rangle_{\mathrm{sym}}`$: the uniform superposition of ordered strings in which label $`x`$ occurs $`n_x`$ times. For the uniform-magnitude source, its coefficient has squared magnitude + +```math +p_n=\frac{t!}{M^t\prod_x n_x!}, +\qquad \sum_xn_x=t. +``` + +The new calculation to perform in this basis is a finite character average: + +```math +\frac1q\sum_{z=0}^{q-1}\omega_q^{(n_x-m_x)z} +=\begin{cases} +1,&n_x\equiv m_x\pmod q,\\ +0,&\text{otherwise}. +\end{cases} +``` + +It follows that a root-phase block retains exactly the occupation coherences with coordinatewise equal residues modulo $`q`$. Continuous independent address phases retain only equal occupations. These two averaging rules are research ingredients applied here; the textbook's symmetric-space framework alone does not establish the persistence law. + +If every $`n_x\lt q`$, no different nonnegative occupation vector of the same total can have those residues. Indeed, writing $`m_x=n_x+qk_x`$ forces every $`k_x\ge0`$, and equal totals force all of them to vanish. Thus the two averaged states have an identical good component. Their remaining weight is bounded by + +```math +\Pr[\text{a label occurs at least }q\text{ times}] +\le \binom tq M^{1-q}. +``` + +Here permutation-invariant vectors and permutation-invariant density operators should be distinguished. A refreshed state can commute with every register permutation without being supported entirely on the symmetric-vector subspace. The worked example gives both kinds of state explicitly. + +## Why trace norm reaches every receiver + +This project defines trace distance with the factor + +```math +D(\rho,\sigma)=\frac12\lVert \rho-\sigma\rVert_1. +``` + +The measurement comparison in §3.1.1 implies that, for any effect $`0\le E\le I`$, + +```math +|\mathrm{Tr}(E\rho)-\mathrm{Tr}(E\sigma)| +\le D(\rho,\sigma). +``` + +For example, this follows from Proposition 3.5 applied to the two-outcome measurement $`\{E,I-E\}`$: the two probability differences are opposite, so their total absolute sum is twice the displayed difference. Any receiver followed by a final yes/no success readout induces such an effect on its entire input. Its ancillas, adaptive operations and memory therefore cannot evade the full-state comparison. + +For a fixed support, replace each root-phase block by its continuous-phase comparison state and telescope across blocks. The [converse](../science/phase-twirl-converse.md) obtains + +```math +D(\Omega_{q,S}^{\mathbf t},\Gamma_S^{\mathbf t}) +\le\min\left\{1, + M^{1-q}\sum_j\binom{t_j}{q}\right\}. +``` + +This is an unconditional comparison of normalized inputs. It does not condition the experiment on a collision and does not restrict the receiver to separate block measurements. + +## Three further steps supplied by the research argument + +The textbook tools need three task-specific additions before they become the theorem. + +1. **A comparison that preserves all input information.** The continuous-phase block is generated exactly by drawing ordinary labels from $`S`$, computing their histogram and coherently preparing its normalized occupation state. This inherited coupon-state simulation retains permutation coherence, including repeated labels; it is not a claim that the block is diagonal in the ordered-string basis. +2. **The posterior for an absent label.** Under a uniform prior on balanced supports, a transcript containing $`d`$ distinct labels leaves a uniform posterior over supports containing them. Every unobserved label then has complement probability $`M/(N-d)`$. No computation on that transcript can improve on its best posterior choice. Together with the trace-distance comparison, this bounds the actual classical output. +3. **A physical route to the matching upper bound.** The inherited coherent power instrument removes the root phases while erasing any residual record that would identify the output label. Its success probability, failed attempts and block tails must all be charged. The established positive-subset diffuser then produces a complement label. General channel and instrument definitions make this a legitimate operation; they do not supply its multiplicity correction or its success formula. + +The [attribution account](../references/related-work.md) identifies the phase-moment, coupon-simulation, power-extraction and positive-diffuser antecedents. The two directions combine to give, for fixed $`q`$, + +```math +Q_q(N,B) +=\Theta_q\!\left( +\min\left\{N, +\max\left\{N^{1-1/q},(N/B)^{q-1}\right\} +\right\}\right). +``` + +The [frontier](../science/persistence-frontier.md) supplies the complete theorem and quantifiers. It counts every acquired specimen, has unconditional success at least $`2/3`$, and allows arbitrary joint processing across known exact $`B`$-call source blocks. The upper bound works for every permitted table sequence; the converse uses independent uniform tables as a hard prior. The basis and finite root alphabet are promises, not learned free resources. + +This is a supplied-copy result. No membership oracle, source preparation program, inverse, reflection or QRAM is supplied. Source construction and receiver gates, storage and synthesis are accounted for separately in [implementation](../science/implementation.md). The bridge and finite example do not establish an end-to-end classical-data runtime advantage, unknown-boundary robustness or the asymptotic proof by numerical inspection. + +[Worked example](WORKED_EXAMPLE.md) · [Source model](../science/model.md) · [Scientific reading route](../science/README.md) diff --git a/learning/WORKED_EXAMPLE.md b/learning/WORKED_EXAMPLE.md new file mode 100644 index 0000000..524969a --- /dev/null +++ b/learning/WORKED_EXAMPLE.md @@ -0,0 +1,218 @@ +# Four labels: the same one-copy state, different two-copy states + +This example uses the density-operator, instrument and symmetric-space language connected to the selected Watrous 2018 anchor in the [bridge](WATROUS_BRIDGE.md). It is a complete finite calculation, not evidence for an asymptotic lower bound or for a sample advantage at this small size. + +Let $`N=4`$, $`M=2`$, $`q=2`$ and, for the calculation, take + +```math +S=\{0,1\},\qquad \bar S=\{2,3\}. +``` + +The receiver is not told $`S`$. Fixing its labels here merely makes the states and output probabilities visible; the displayed receiver operations do not depend on that choice. + +## 1. Positive input and every diffuser output + +The positive subset state and known uniform state are + +```math +|S\rangle=\frac{|0\rangle+|1\rangle}{\sqrt2}, +\qquad +|u_4\rangle=\frac{|0\rangle+|1\rangle+|2\rangle+|3\rangle}{2}. +``` + +The established balanced-subset diffuser is + +```math +D_4=2|u_4\rangle\langle u_4|-I +=\frac12 +\begin{pmatrix} +-1&1&1&1\\ +1&-1&1&1\\ +1&1&-1&1\\ +1&1&1&-1 +\end{pmatrix}. +``` + +Direct multiplication gives + +```math +D_4|S\rangle=\frac{|2\rangle+|3\rangle}{\sqrt2}. +``` + +For signs, a global sign has no effect on a density operator, so the independent uniform sign prior can be written as a uniform relative sign $`r\in\{+1,-1\}`$ with + +```math +|\psi_r\rangle=\frac{|0\rangle+r|1\rangle}{\sqrt2}. +``` + +The plus input is the positive state above. The minus input is orthogonal to $`|u_4\rangle`$, hence $`D_4|\psi_{-1}\rangle=-|\psi_{-1}\rangle`$. All four measurement probabilities after diffusion are: + +| Input to the diffuser | Output 0 | Output 1 | Output 2 | Output 3 | Complement probability | +| --- | --- | --- | --- | --- | --- | +| $`\lvert\psi_{+1}\rangle`$ | $`0`$ | $`0`$ | $`1/2`$ | $`1/2`$ | $`1`$ | +| $`\lvert\psi_{-1}\rangle`$ | $`1/2`$ | $`1/2`$ | $`0`$ | $`0`$ | $`0`$ | +| Uniform hidden sign prior | $`1/4`$ | $`1/4`$ | $`1/4`$ | $`1/4`$ | $`1/2`$ | + +Thus applying the positive-state protocol directly to an unknown-sign specimen does not preserve its guarantee. The diffuser and the original positive-state complement task are inherited from Benedetti, Buhrman and Weggemans; see [related work](../references/related-work.md). + +## 2. Compute the one-copy average + +The two pure-state matrices have opposite off-diagonal entries. Their average is + +```math +\rho_S +=\frac12|\psi_{+1}\rangle\langle\psi_{+1}| + +\frac12|\psi_{-1}\rangle\langle\psi_{-1}| +=\frac12|0\rangle\langle0| + +\frac12|1\rangle\langle1|. +``` + +Every individual emission has this averaged state whether its hidden signs will be reused or independently refreshed. That agreement does not fix the joint state of two emissions. + +## 3. Freeze the signs for two calls + +Suppose both calls belong to one known phase block, so $`B\ge2`$ and the same $`r`$ is used. Write + +```math +|a\rangle=\frac{|00\rangle+|11\rangle}{\sqrt2}, +\qquad +|b\rangle=\frac{|01\rangle+|10\rangle}{\sqrt2}. +``` + +Then + +```math +|\psi_r\rangle^{\otimes2} +=\frac12\bigl(|00\rangle+r|01\rangle+r|10\rangle+|11\rangle\bigr) +=\frac{|a\rangle+r|b\rangle}{\sqrt2}. +``` + +Averaging the fixed sign gives + +```math +\Omega_{2,S}^{(2)} +=\frac12|a\rangle\langle a|+\frac12|b\rangle\langle b| +=\frac14 +\begin{pmatrix} +1&0&0&1\\ +0&1&1&0\\ +0&1&1&0\\ +1&0&0&1 +\end{pmatrix}, +``` + +where the displayed basis is $`00,01,10,11`$; all other two-register basis entries have zero weight. In particular, + +```math +\langle00|\Omega_{2,S}^{(2)}|11\rangle=\frac14. +``` + +The coherence survives because its phase factor is the square of a sign. In occupation language, $`(2,0)`$ and $`(0,2)`$ have the same coordinatewise residues modulo two. + +## 4. Refresh the signs at every call + +If the tables are independently redrawn for each call, the joint average is + +```math +\rho_S\otimes\rho_S +=\frac14\left( +|00\rangle\langle00|+|01\rangle\langle01| ++|10\rangle\langle10|+|11\rangle\langle11| +\right). +``` + +Now the $`|00\rangle\langle11|`$ entry is zero. The $`|01\rangle\langle10|`$ entry is also zero: the two source calls no longer share a relative sign. + +The frozen state is supported on symmetric vectors. The refreshed state commutes with swapping the registers, but has weight $`1/4`$ on the antisymmetric vector $`(|01\rangle-|10\rangle)/\sqrt2`$. Commuting with a permutation is therefore weaker than being supported on its invariant-vector subspace. + +As an exact comparison at this fixed support, + +```math +D\!\left(\Omega_{2,S}^{(2)},\rho_S^{\otimes2}\right)=\frac12. +``` + +Indeed, their difference has four eigenvalues $`1/4,1/4,-1/4,-1/4`$ on the displayed subspace. This distinguishes joint states; it is not a complement-sampling lower bound with an unknown support. + +For comparison, phases drawn uniformly from the circle and then frozen for the pair give + +```math +\Gamma_S^{(2)} +=\frac14|00\rangle\langle00| + +\frac12|b\rangle\langle b| + +\frac14|11\rangle\langle11|. +``` + +It retains the permutation coherence inside $`|b\rangle`$ but loses the coherence between different occupations. Thus continuous-phase averaging is not the same operation as independently dephasing every emission. In this example $`D(\Omega_{2,S}^{(2)},\Gamma_S^{(2)})=1/4`$, below the converse's collision bound $`\binom22/M=1/2`$. + +## 5. Recover coherently: two filters with different probabilities + +The inherited general-batch instrument in [coherent recovery](../science/coherent-recovery.md) has + +```math +C_2=\frac{3!}{1!}=6. +``` + +For $`b=q=2`$, the only residual occupation is zero, so its success operator reduces to + +```math +K_0=\frac1{\sqrt3}\sum_{x=0}^{3}|x\rangle\langle xx|. +``` + +For either fixed sign, + +```math +K_0|\psi_r\rangle^{\otimes2} +=\frac{|0\rangle+|1\rangle}{2\sqrt3} +=\frac1{\sqrt6}|S\rangle. +``` + +Its success probability is therefore + +```math +p_2=\frac{(2)_2}{C_2M} +=\frac{2}{6M} +=\frac1{3M} +=\frac16. +``` + +On the successful branch, the known diffuser gives a complement label with certainty. The success record reveals no surviving label, so the relative coherence needed by the diffuser remains. + +There is also a stronger filter for this particular two-copy case: + +```math +A=\sum_{x=0}^{3}|x\rangle\langle xx|, +\qquad +A^\dagger A=\sum_x|xx\rangle\langle xx|\le I. +``` + +It succeeds with probability $`1/M=1/2`$ and has the same normalized powered output on frozen-sign inputs. It is the unscaled coherent equality filter, not the $`K_0`$ operator of the general-batch construction. Hence $`1/M`$ must not be substituted into that construction's stated success formula. Reading both labels and recording which equal label occurred would also have equality probability $`1/M`$, but would destroy the coherent recovered state. + +On independently refreshed inputs, the coherent equality filter succeeds with the same probability, yet its average successful output is $`\rho_S`$, not $`|S\rangle\langle S|`$. Equal labels alone are insufficient; persistence is what makes the signs cancel coherently. + +## 6. Include failure branches and ordinary samples + +Complete the inherited $`K_0`$ instrument with failure, and on failure return an independent uniform label. This gives a hard two-copy protocol with every run returning an answer. Its four output probabilities are + +```math +(\Pr[Y=0],\Pr[Y=1],\Pr[Y=2],\Pr[Y=3]) +=\left(\frac5{24},\frac5{24},\frac7{24},\frac7{24}\right). +``` + +The unconditional complement probability is $`7/12`$, not one. Certainty applies only after the success flag. + +Three fresh two-copy attempts using the inherited operator, with a uniform label only if all fail, have success + +```math +1-\frac12(1-1/6)^3 +=\frac{307}{432}>\frac23. +``` + +For exact blocks of length $`B=2`$, this has a hard six-copy cap. For $`B=3`$, each failed block leaves one specimen to acquire and discard before the next attempt, so the cap becomes eight copies. Those tails count even though the receiver does not use them. At $`B=1`$, no pair of consecutive calls has the promised frozen signs, so this recovery argument cannot be applied. + +The small example also makes clear why a successful coherent branch is not itself a copy advantage. Measure one ordinary specimen and return a uniformly chosen different label. For every balanced support, two of the remaining three labels lie outside $`S`$, giving unconditional success exactly $`2/3`$. For the fixed support used here its output probabilities are $`(1/6,1/6,1/3,1/3)`$. + +Under a uniform prior over the six balanced supports, this is also the optimal one-copy averaged success: after seeing one supported label, all three unobserved labels have posterior complement probability $`2/3`$. Two ordinary draws followed by a uniform unobserved label give success $`5/6`$: equal observations occur with probability $`1/2`$ and give success $`2/3`$, while distinct observations reveal the whole support and give success one. + +These finite values illustrate source correlations, coherent cancellation, unconditional success and charged acquisition. The fixed-$`q`$, growing-$`N`$ optimal law requires the full [frontier proof](../science/persistence-frontier.md). The known basis, balanced size and exact block boundaries remain assumptions; no hidden-set oracle, source inverse, reflection, preparation program or free data loading has been introduced. + +[Watrous bridge](WATROUS_BRIDGE.md) · [Source model](../science/model.md) · [Converse](../science/phase-twirl-converse.md) · [Recovery and schedule](../science/coherent-recovery.md) diff --git a/llms.txt b/llms.txt index 30d3545..7924053 100644 --- a/llms.txt +++ b/llms.txt @@ -16,6 +16,13 @@ Structured bibliography: references/references.bib Finite checks and their limits: verification/README.md Reproduction command: python3 verification/check_science.py +Educational route: learning/README.md +Single educational anchor: John Watrous, The Theory of Quantum Information (2018), official author page https://cs.uwaterloo.ca/~watrous/TQI/ +Textbook section translation and missing proof steps: learning/WATROUS_BRIDGE.md +Small source-to-output calculation: learning/WORKED_EXAMPLE.md +Exactly 30 intrinsic milestones: learning/MILESTONES.md (M01-M12 background and preliminaries; M13-M30 narrative and innovation; local-to-global progression in each phase) +Learning structure and navigation command: python3 verification/check_learning.py + Keep these distinctions: fixed q, balanced support, known exact B-call blocks, supplied specimens only, every failed/discarded acquisition charged, unconditional classical output. No source inverse/reflection, membership oracle, hidden phase program or QRAM is supplied. B is not a receiver-memory cap. Refreshed dephasing requires vanishing pairwise phase correlations across different addresses, satisfied by independent uniform root phases. Persistent and refreshed sources share one-copy averages but different many-copy states. The power extractor, coupon simulator, phase averaging, birthday method and positive complement swapper are inherited and attributed. The new account concerns their matching classical-output persistence consequence. Do not infer a coherent state-transformation lower bound from the sampling theorem or the reverse. Do not infer end-to-end classical-data speedup, generic noise robustness, a hardware crossover, growing-q uniformity or exhaustive priority. diff --git a/science/coherent-recovery.md b/science/coherent-recovery.md index 5ac2b17..c47ccbc 100644 --- a/science/coherent-recovery.md +++ b/science/coherent-recovery.md @@ -34,7 +34,7 @@ Define C_q=\frac{(2q-1)!}{(q-1)!}. ``` -For each residual occupation vector $`z`$ satisfying $`|z|=b-q`$ and $`0\le z_x\frac23, \qquad -0\le\eta<\frac{\pi}{2}. +0\le\eta\lt \frac{\pi}{2}. ``` Within each source block, permit diff --git a/science/persistence-frontier.md b/science/persistence-frontier.md index 945769b..9165aca 100644 --- a/science/persistence-frontier.md +++ b/science/persistence-frontier.md @@ -13,6 +13,8 @@ Let $`Q_q(N,B)`$ be the minimum hard total-copy budget for unconditional complem **Theorem 1 — phase-persistence law.** There are positive constants depending only on fixed $`q`$ such that, uniformly over integers $`B\ge1`$, +**Equation (1).** + ```math \boxed{ Q_q(N,B) @@ -27,7 +29,6 @@ N^{1-1/q}, \right\} \right). } -\tag{1} ``` The statement is asymptotic in $`N`$. The constants also depend on the chosen fixed success threshold if that threshold is changed. Small finite dimensions do not alter the order. @@ -56,7 +57,9 @@ These are operational scales in one supplied-copy task. They are not claims that ## 3. Proof route: lower bound -The [phase-twirl converse](phase-twirl-converse.md) proves the following finite bound. For a receiver using at most $`T\frac23. diff --git a/science/phase-twirl-converse.md b/science/phase-twirl-converse.md index 395601a..33fb6a8 100644 --- a/science/phase-twirl-converse.md +++ b/science/phase-twirl-converse.md @@ -7,7 +7,7 @@ The two inherited ideas are phase averaging in the symmetric occupation basis an Throughout, $`N=2M`$, $`|S|=M`$, the integer $`q\ge2`$ is fixed, and phase tables persist for exactly $`B`$ source calls as specified in the [model](model.md). All trace distances use ```math -D(\rho,\sigma)=\frac12\|\rho-\sigma\|_1. +D(\rho,\sigma)=\frac12\lVert \rho-\sigma\rVert_1. ``` ## 1. One block in the occupation basis @@ -70,9 +70,10 @@ n_x\equiv m_x\pmod q\;\forall x}} \sqrt{p_np_m}\, |n\rangle_{\mathrm{sym}} \langle m|_{\mathrm{sym}}. -\tag{1} ``` +Equation (1). + By comparison, independent uniform continuous phases at each address give ```math @@ -81,9 +82,10 @@ By comparison, independent uniform continuous phases at each address give \sum_n p_n |n\rangle_{\mathrm{sym}} \langle n|_{\mathrm{sym}}. -\tag{2} ``` +Equation (2). + The phase vector is drawn once for the whole block in both equations. Equation (2) is generally not diagonal in the ordered computational basis: permutations of the same histogram remain coherent. ## 2. Which coherences require a repeated label? @@ -93,17 +95,17 @@ Call an occupation vector good when every label occurs fewer than $`q`$ times. L ```math \Pi_{\mathrm{good}} = -\sum_{\max_x n_x