--- id: QuantumComplementaryChannelContract language: en section: quantum-physics source: Verification/QuantumComplementaryChannelContract.lean source_sha256: 2ca2ee2960200b0e3ec2393ff7f7fdf7b5cb385998a7385f34a5d9c8e985c39a novelty: not-assessed status: reviewed --- # Complementary quantum channel and environment state [Section](../quantum-physics/README.md) · [Lean](../../Verification/QuantumComplementaryChannelContract.lean) ## Input and construction The input is a supplied `KrausChannel d e r` from module 175. The matrices K_j have shape e by d and satisfy sum K_j† K_j = I_d. The module reuses the exact Stinespring matrix and dilation of module 177, with product index **output system × environment**. Dimensions d, e, r need not coincide. `partialTrace_system M i j = sum_a M (a,i) (a,j)` traces out the first index. The existing `partialTrace` traces out the second index. The new function is a finite-index definition for this layout, not a change to the existing API. `complementary_channel K X = partialTrace_system (dilation K X)`. `complementary_entry` proves its (i,j) entry equals Tr(K_i X K_j†), with this precise order of conjugation, for every complex matrix X. `complementary_operators K a j b = K_j a b`. Their completeness follows by exchanging the two finite sums in the original completeness equation. `complementary_kraus K : KrausChannel d r e` therefore represents the environment output with e Kraus operators. `complementary_eq_apply` identifies this channel with the partial-trace construction. Complete positivity is obtained directly from module 175's finite block-amplification definition, not assumed or replaced by ordinary positivity. The module also proves linearity, trace preservation and preservation of density matrices through the same API. `simultaneous_reductions` obtains both channels from the same joint matrix. For arbitrary X, this is just a complex matrix. Density inputs give density joint matrices by module 177; mixed inputs need not become pure. All identities are finite-dimensional, including algebraic zero-dimension cases. No trace-one state is asserted to exist in a zero-dimensional space. ## Explicit changes of environment For W of shape r' by r, `environment_lift W = I_e tensor W` and `change_environment W M = (I_e tensor W) M (I_e tensor W)†`. The block identity proves that each fixed-system block changes by M_ab -> W M_ab W†. Summing diagonal blocks proves partialTrace_system(change_environment W M) = W * partialTrace_system(M) * W†. This covariance identity holds for arbitrary W and arbitrary M. The separate hypothesis W†W=I_r is necessary for the proved invariance of the other reduction: `main_environment_invariant` uses cyclicity of the trace of each block. `changed_stinespring_isometry` proves that V'=(I_e tensor W)V remains an isometry. `changed_kraus` reconstructs an actual `KrausChannel d e r'` from its entries. `changed_stinespring` and `changed_dilation` identify the constructions exactly. `changed_main_channel` proves Phi'=Phi; `changed_complementary_channel` proves Phi_c'(X)=W Phi_c(X) W†. W may be rectangular. This is a theorem about a GIVEN isometric transformation, not existence of an isometry relating arbitrary representations or a minimality/uniqueness theorem. ## Examples and regression evidence For a unitary channel with a one-dimensional environment the complement is Tr(X) I_1; for density inputs it is the fixed state I_1. The reset channel to a fixed basis state has complement X itself, including unequal input/output dimensions. Computational-basis dephasing has complement diag(X_ii). For qubit |+> and |->, BOTH reduced channels give the same outputs, while the joint dilations differ. `correlations_retain_distinction` proves all three statements together. The environment alone therefore need not contain every distinction absent from the system. The joint correlations can retain it. Appending a zero Kraus operator leaves the main channel unchanged by module 177. The new complementary matrix has the old environment as its leading block and an exactly zero last row and column; three separate theorems establish this. The new live regressions check a rectangular 3-by-2 environment isometry with an imaginary phase, non-Hermitian input, density normalization, unequal reset dimensions, mixed dephasing output, repeated padding, and a zero non-isometry that fails invariance. These are exact Lean kernel checks over complex matrices, not floating-point simulations or an executable complex-number engine. ## Clean Water and provenance The complementary channel depends on the chosen Kraus representation. No claim is made that all information missing from the main output is accessible in the environment marginal alone. General isometric uniqueness, minimal environment, unitary extension, infinite dimensions, quantum capacity, general recovery maps, numerical certification and physical implementation remain outside scope. Full Bulletproofs protocol extraction/security (beyond certified openings), Tendermint reference acquisition/slashing, Python correspondence, JSON/SHA-256 and SimLab remain separate obligations. Mathematical novelty is not assessed. Public gamma release v0.5.79 exports private snapshot 6efc674f70eb0181cd1c5667f919bcf0826cdd21. The previous public scoped release is 9d95f47fb7970af648f3c6b31977e627283dcca7 (v0.5.78). The last fully validated PUBLIC baseline remains b3410720b9af49c2fb70d15d65d3d012091cbf0b (v0.5.73). Public-tree checks are recorded in [the release report](../../docs/release-v0.5.79.en.md). Private evidence is not substituted for public validation; no new full-project verify-all or full unittest result is claimed. The permitted axioms remain propext, Classical.choice and Quot.sound. Background: [Watrous, The Theory of Quantum Information](https://cs.uwaterloo.ca/~watrous/TQI/).