silva_quantum_deq Reproduction Dossier

measured quantum-circuit equilibrium. This dossier connects the source mechanism to its SILVA implementation, compact evidence, replaceable components, and source-scale route. Existing tests and notebooks remain the executable authority.

Evidence boundary

The mechanism is compact-verified in the package suite. The final source-scale stage remains planned until the cited data, complete optimization budget, checkpoints, and evaluation protocol have actually run.

Identity and Sources

Field	Value
Domain	vision and generation
Task contract	image or vector input -> injected measured state -> task output
Source relation	paper-adaptation
References	[90]
Repositories	https://github.com/martaskrt/qdeq
Editable scale plan	experiments/reproduction/configs/silva_quantum_deq.json

Governing Equation

The domain-level state contract is

\[ z^\star=T_\theta(z^\star;\mathcal E(x),c),\qquad \widehat y=\mathcal D_\psi(z^\star). \]

The implementation registry specializes it operationally as

z_star=Measure(U_theta(Encode(z_star+S(x)))); y_hat=Q(z_star)

Define the root residual

\[ R_\theta(z;x)=z-T_\theta(z;x). \]

At a regular equilibrium, differentiating \(R_\theta(z^\star;x)=0\) gives

\[ \frac{\partial z^\star}{\partial x} = \left(I-\frac{\partial T_\theta}{\partial z}\right)^{-1} \frac{\partial T_\theta}{\partial x}. \]

This identity explains why the forward residual, the conditioning derivative, and the adjoint linear solve must be diagnosed separately from the task metric.

What Is Preserved

• feature injection, repeated quantum-circuit measurement, and fixed-point solving
• direct, warmup, and implicit training routes with Jacobian regularization

What Can Be Replaced

Each item below is an explicit control rather than an undocumented modification:

• replace the input adapter, circuit, measurement map, readout, or solver
• switch between finite tied steps, warmup, and implicit differentiation
• wire count
• circuit depth
• root-solver budget
• Jacobian penalty frequency

Constructor and Shape Contract

silva_quantum_deq(input_dim: 'int', output_dim: 'int', *, n_qubits: 'int' = 4, input_adapter: 'nn.Module | None' = None, circuit: 'nn.Module | None' = None, readout: 'nn.Module | None' = None, mode: 'QuantumExecutionMode' = 'implicit', direct_steps: 'int' = 10, warmup_steps: 'int' = 0, config: 'SolverConfig | None' = None) -> 'None'

The transition must preserve the declared equilibrium-state shape even when the encoder, branch operators, constraints, solver, and readout are replaced. Test the transition by itself before testing the complete root solve.

Progressive Experiment Ladder

1. Equation and tensor contract

Objective: Make the state, conditioning variables, operator, and readout explicit.

Procedure:

• Write and evaluate the family equation: z_star=Measure(U_theta(Encode(z_star+S(x)))); y_hat=Q(z_star)
• Declare every tensor axis, boundary, mask, graph, or physical unit.
• Check the transition output has exactly the same shape as the equilibrium state.

Acceptance checks:

• finite transition values
• shape-preserving state update
• all conditioning variables affect the intended branch

Evidence target: contract-verified.

2. Primitive mechanism reconstruction

Objective: Build the retained source mechanism from replaceable modules.

Procedure:

• feature injection, repeated quantum-circuit measurement, and fixed-point solving
• direct, warmup, and implicit training routes with Jacobian regularization

Acceptance checks:

• primitive modules expose trainable parameters and gradients
• mechanism-specific invariance or constraint check passes
• direct transition evaluation is deterministic under a fixed seed

Evidence target: compact-verified.

3. Public abstraction equivalence

Objective: Verify that the assembled family evaluates the same transition as its primitives.

Procedure:

• Copy the primitive module parameters into the public family constructor.
• Evaluate one transition and one complete equilibrium with identical inputs.
• Compare outputs, residuals, and parameter gradients with declared tolerances.

Acceptance checks:

• transition outputs agree
• equilibrium residual is finite and decreases
• primitive and assembled gradients agree on the compact case

Evidence target: compact-verified.

4. Compact real or analytic task

Objective: Exercise training, evaluation, diagnostics, and serialization end to end.

Procedure:

• The QDEQ lab uses seeded normalized feature directions, exact statevector measurements, and a compact binary classification target.
• SILVAQuantumImageFilter verifies the source 28x28 image-to-circuit shape contracts before a licensed dataset is introduced.

Acceptance checks:

• record classification accuracy
• record fixed-point residual and iterations
• record circuit evaluations
• record Jacobian penalty and gradient variance
• checkpoint reload reproduces the recorded prediction
• result record contains data and configuration fingerprints

Evidence target: compact-verified.

5. Official-data subset

Objective: Validate the complete source data path before spending the full budget.

Procedure:

• Acquire one declared image benchmark, preserve its official split, and reproduce the source image filter, class subset, encoding, wire count, and circuit seed.
• Freeze preprocessing, split logic, metric code, and checkpoint format.
• Run a deterministic subset large enough to expose batching and memory failures.

Acceptance checks:

• dataset receipt and checksum are stored
• resume and evaluation paths reproduce the same subset metric
• memory and runtime are measured rather than estimated

Evidence target: subset-verified.

6. Source-scale reproduction or declared extension

Objective: Run the cited protocol, or change it explicitly as a SILVA extension.

Procedure:

• Acquire one declared image benchmark, preserve its official split, and reproduce the source image filter, class subset, encoding, wire count, and circuit seed.
• Match the fixed and trainable gate sequences, measurement/interpolation rule, direct warmup, implicit-solver budget, backward rule, and Jacobian regularization schedule.
• Report task accuracy, residual, iterations, circuit evaluations, gradient variance, wall time, memory, and shots or exact-statevector setting against direct and classical baselines.
• source dataset split, wire count, encoding, circuit seed, solver budgets, schedule, and task metric

Acceptance checks:

• all required artifacts are archived
• reported metrics use the cited evaluation protocol
• every architectural or training deviation is listed
• claims match the achieved evidence status

Evidence target: planned.

Data, Access, and Storage

Candidate datasets:

• MNIST-4
• MNIST
• Fashion-MNIST
• CIFAR-10
• compact exact-statevector classification

Authoritative routes:

• https://arxiv.org/abs/2410.23940
• https://github.com/martaskrt/qdeq
• https://yann.lecun.com/exdb/mnist/
• https://github.com/zalandoresearch/fashion-mnist
• https://www.cs.toronto.edu/~kriz/cifar.html

Access obligations:

• MNIST, Fashion-MNIST, and CIFAR-10 have established public acquisition routes under their stated terms.
• MNIST-4 is a declared four-class subset rather than a separate archive; preserve the chosen classes, split indices, resizing, channel conversion, and normalization.
• Record the circuit backend and version, encoding, wire ordering, fixed-circuit seed, gate pattern, measurement type, and shot count or exact-statevector setting.

Storage planning:

• The public image datasets fit comfortably within a few gigabytes, but exact statevector work memory scales as batch size times 2^wires complex amplitudes.
• Shot-based backends additionally scale with samples * equilibrium evaluations * measured observables * shots; record this separately from host-side tensors.
• Store image split indices, filtered features, circuit parameters, solver traces, and checkpoints independently so preprocessing can be audited without duplicating raw data.

Preprocessing record:

• record dataset version, split, normalization, shape convention, and seed
• preserve masks, graph indices, boundaries, or physical units required by the domain

Metrics and Current Evidence

Required metrics:

• classification accuracy
• fixed-point residual and iterations
• circuit evaluations
• Jacobian penalty and gradient variance
• wall time, memory, and shot count

This family is verified through its listed mechanism tests and executed notebook. It is not included in a same-task comparison when another family does not share its input, state, output, and loss contract. The absence of a comparison row is therefore a scope decision, not missing implementation evidence.

Executed notebook paths:

• notebooks/package_api/50_silva_quantum_deq.ipynb
• notebooks/package_api/51_equilibrium_expansion_atlas.ipynb

Mechanism tests:

• tests/test_quantum_equilibria.py

Compact Defaults

Option	Value
tier	'smoke'

Full Defaults

Option	Value
tier	'full'

Defaults establish a starting budget; the cited source protocol takes precedence whenever reproduction is the claim.

Source-Scale Checklist

• Acquire one declared image benchmark, preserve its official split, and reproduce the source image filter, class subset, encoding, wire count, and circuit seed.
• Match the fixed and trainable gate sequences, measurement/interpolation rule, direct warmup, implicit-solver budget, backward rule, and Jacobian regularization schedule.
• Report task accuracy, residual, iterations, circuit evaluations, gradient variance, wall time, memory, and shots or exact-statevector setting against direct and classical baselines.

Benchmark-specific requirements:

• source dataset split, wire count, encoding, circuit seed, solver budgets, schedule, and task metric

Required archived artifacts:

• machine-readable model and solver configuration
• dataset receipt with source revision, split, license, and checksum
• preprocessing and normalization record
• seeded training and evaluation log
• checkpoint and optimizer-resume state for trained experiments
• task metrics and equilibrium diagnostics in a machine-readable result
• runtime, peak-memory, device, precision, and dependency record
• declared deviations from the cited protocol

Reporting Rule

Report the achieved evidence status, not the intended one. A compact or subset run may validate the implementation and data path, but only a completed cited protocol supports a source-scale reproduction statement. Modified operators are valuable SILVA extensions when every deviation is named and measured.

Where to Go Next

Question	Page
Where are all family dossiers?	Family Dossier Index
How is a custom family assembled?	Advanced Extension Handbook
How are experiment stages represented in the API?	Research-Depth API
Which lab inspects every dossier?	Family Dossier Lab