Alphabeta Math
Physics theoremStatement: AI-adaptedProof: AI-generatedPipeline-generatedverified 2026-10-22 (gpt-6.1-sol)
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

Product-state reduced evolution is a channel

Physical scope

Exact reduced maps of finite-dimensional system and environment models with a fixed product initial assignment and unitary joint evolution. The theorem permits neither an initial-correlation extension nor a continuum bath or Markov limit; its finite example is a conditional model calculation, not an experiment.

Statement

In the finite product-state model, fix any t≥0 and define the complex-linear map Φt:Md(C)→Md(C) by Φt(X)=Tr⁡E ⁣(Ut(X⊗η)Ut∗). Then Φt is a finite quantum channel. The family t↦Φt need not obey the semigroup law and need not be CP-divisible. The terms “semigroup law” and “CP-divisible” have the definitions of Finite channel families, semigroups, and CP-divisibility.

Facts & Assumptions

Given: Integers d,m≥1, time t≥0, system and environment density matrices ρ∈Md(C) and η∈Mm(C), and a unitary Ut on Cd⊗Cm with the system factor first.

[A1]

Finite system-environment and Markov model assignments: η is fixed independently of every allowed input ρ, the joint initial state is ρ⊗η, and the reduced map is Φt(X)=Tr⁡E(Ut(X⊗η)Ut∗); the parameter t indexing Ut is elapsed physical time measured in seconds.

[F1]

Partial trace over a finite tensor factor: in the ordered product basis, (Tr⁡ER)αβ=∑jR(α,j),(β,j).

[F2]

The partial trace is characterized by the trace pairing and is basis-independent: the finite partial trace is characterized by the trace pairing and its coordinate map is independent of the chosen environment orthonormal basis.

[F3]

Finite matrix spectral and state calculus: every finite positive matrix has an orthonormal eigenbasis with nonnegative eigenvalues, and its trace is the sum of its diagonal entries in any orthonormal basis.

[F4]

Finite completely positive trace-preserving maps: a finite channel is a complex-linear map that is completely positive on every finite ancilla extension and trace-preserving on every matrix input.

[F5]

Complete finite Choi-Kraus equivalence: a finite map of Kraus form Ψ(X)=∑aKaXKa∗ is completely positive, and it is trace-preserving exactly when ∑aKa∗Ka=I.

[F6]

Finite channel families, semigroups, and CP-divisibility: CP-divisibility requires a finite channel Vt,s for every s≤t with Φt=Vt,s∘Φs; the semigroup law requires Φt+s=Φt∘Φs and Φ0=id⁡.

[F7]

The complex exponential by its power series: exp⁡z=∑n=0∞zn/n! and exp⁡0=1.

[F8]
[F9]

exp⁡(x+iy)=ex(cos⁡y+isin⁡y), ∣exp⁡(x+iy)∣=ex, and eiπ+1=0: for real θ, exp⁡(iθ)=cos⁡θ+isin⁡θ and ∣exp⁡(iθ)∣=1.

[F10]

t↦(cos⁡t,sin⁡t) is a bijection from [0,2π) onto the real unit circle: there exists θ∈[0,2π) with (cos⁡θ,sin⁡θ)=(0,1).

Proof

technique · finite Kraus construction followed by an explicit two-qubit counterexample
1.1givenA1F1F2F3constructalgebra

By [F3], choose an orthonormal environment eigenbasis (∣k⟩)k=1m with η=∑k=1mpk∣k⟩⟨k∣, pk≥0, and ∑kpk=1. Use the fixed system basis from the model and its ordered product with this environment basis. By [F2], changing the environment basis does not change the partial-trace map in [F1]. For environment indices j,k define the system matrix Kjk by (Kjk)αβ:=pk(Ut)(α,j),(β,k). Expanding X⊗η, conjugating by Ut, and applying the coordinate formula [F1] gives, entry by entry, Φt(X)=∑j,k=1mKjkXKjk∗ for every X∈Md(C).

1.2givenA1F1F7F9constructalgebra

For a two-qubit instance take Z=diag⁡(1,−1), g>0 in s−1, Htot=ℏg Z⊗Z with ℏ in joule-seconds, and η=∣+⟩⟨+∣ where ∣+⟩=(∣0⟩+∣1⟩)/2 in the chosen environment Z eigenbasis. Put A=Z⊗Z and Pb=(I+bA)/2 for b=±1; direct multiplication gives PbPb′=δbb′Pb, P++P−=I, and A=∑b=±1bPb. Define the matrix exponential for this instance by the power series Ut:=∑n=0∞(−igt)nAn/n!. Since An=∑bbnPb, the series converges entrywise by the scalar exponential definition [F7] and equals ∑be−ibgtPb=e−itHtot/ℏ. In the ordered environment Z eigenbasis this is ∑b=±1Vb(t)⊗∣b⟩⟨b∣, where Vb(t)=diag⁡(e−ibgt,eibgt). By [F9], each scalar phase has modulus one, so Ut is unitary and U0=I. The partial trace in [F1] therefore gives Φt(X)=12(V+1(t)XV+1(t)∗+V−1(t)XV−1(t)∗), so for X=(azwc) the diagonal entries are unchanged and the off-diagonal entries are multiplied by κ(t)=12(e−2igt+e2igt)=cos⁡(2gt).

1.3F7F8F9F10algebra

By [F10], choose θ∈[0,2π) with cos⁡θ=0 and sin⁡θ=1; hence θ>0. Set s=θ/(2g)>0. Euler's identity [F9] gives eiθ=i, and [F7]–[F8] give e−iθ=(eiθ)−1=−i. Applying [F8] again yields κ(s)=0 and κ(2s)=12(e−2iθ+e2iθ)=−1.

2.1givenF4F5step 1.1algebra

For system basis indices β,δ, unitarity of Ut gives (∑j,kKjk∗Kjk)βδ=∑k=1mpk∑j=1m∑α=1d(Ut)(α,j),(β,k)‾(Ut)(α,j),(δ,k)=∑kpkδβδ=δβδ. Thus ∑j,kKjk∗Kjk=Id, so [F5] makes the Kraus map in step 1.1 completely positive and trace-preserving; by [F4], Φt is a finite channel.

3.1F3F6step 1.2step 1.3step 2.1algebra∎

Let X ⁣P=(0110) and ρ±=(I±X ⁣P)/2. Since X ⁣P=X ⁣P∗ and X ⁣P2=I, each ρ± is a projection; since Tr⁡X ⁣P=0, each has trace one. Steps 1.2 and 1.3 give Φs(ρ+)=Φs(ρ−)=I/2, while Φ2s(ρ+)=ρ− and Φ2s(ρ−)=ρ+. Consequently Φ2s(ρ+)≠Φs(Φs(ρ+))=I/2, so the semigroup law in [F6] fails. If a channel V2s,s satisfied Φ2s=V2s,s∘Φs, it would send the equal matrices Φs(ρ+) and Φs(ρ−) to equal outputs, contradicting the two distinct outputs at 2s. By step 2.1, the family consists of channels, so this contradiction proves it is not CP-divisible by [F6].

Remarks

The two-qubit recurrence is a conditional prediction of a finite unitary model. It is not a reported observation and does not establish a physical Markov limit.

Depends on

Dependency tree · 2 levels

Sources