AFTD Auto-Formalizing Theoretical Domains

ppt_qutrit_bell_state_partial_transpose_invariant

theoremverified

The explicit rank-four two-qutrit source is invariant under partial transpose on its first qutrit.

Statement

theorem ppt_qutrit_bell_state_partial_transpose_invariant :
  ∀ a b c d : Fin 3,
    ppt_qutrit_bell_state (a,b) (c,d) = ppt_qutrit_bell_state (c,b) (a,d)
Source
Source-state helper for OP-693; see https://arxiv.org/abs/1405.4502.
Verified
12 Nov 2026
Axioms
Classical.choiceQuot.soundpropext
Built on 2
  • ppt_qutrit_bell_stateThe explicit two-qutrit positive-partial-transpose state in the Bell-violation example.
  • ppt_qutrit_bell_vectorFour explicit real vectors in two-qutrit space used in the PPT Bell-violation example.
Used by 3

The proof, explained

The theorem asserts that the rank-four two-qutrit density matrix is invariant under partial transposition on its first qutrit, meaning its $((a, b), (c, d))$-entry equals its $((c, b), (a, d))$-entry for all $a, b, c, d \in \{0, 1, 2\}$. The proof first establishes algebraic identities for the square roots $\sqrt{2}, \sqrt{3}, \sqrt{131}$ appearing in the state definition, along with the non-zeroness of $\sqrt{2}$ and $\sqrt{3}$. It then introduces a helper lemma that exhaustively verifies all $3^4 = 81$ entry pairs using `fin_cases`, expanding the definition of the state as a sum of four Bell projectors and simplifying each case to equality with automated field and ring normalization. Finally, the main theorem is closed directly by applying this helper lemma.

  1. The square roots $\sqrt{2}, \sqrt{3}, \sqrt{131}$ satisfy $(\sqrt{2})^2 = 2$, $(\sqrt{3})^2 = 3$, and $(\sqrt{131})^2 = 131$ in $\mathbb{C}$, and both $\sqrt{2} \ne 0$ and $\sqrt{3} \ne 0$ in $\mathbb{C}$.

    For each radicand $x \in \{2, 3, 131\}$, non-negativity $0 \le x$ is verified by `norm_num`, so `Real.sq_sqrt` shows $(\sqrt{x})^2 = x$ in $\mathbb{R}$, which is cast to $\mathbb{C}$ via `exact_mod_cast`. Strict positivity $0 < 2$ and $0 < 3$ implies $\sqrt{2} > 0$ and $\sqrt{3} > 0$ by `Real.sqrt_pos.2`, hence non-zero by `ne_of_gt`, which is likewise cast to $\mathbb{C}$.

    lines 10–14

    uses Real.sq_sqrt, ne_of_gt, Real.sqrt_pos.2

      have h2 : (Real.sqrt 2 : ℂ)^2=2 := by exact_mod_cast Real.sq_sqrt (show (0:ℝ)≤2 by norm_num)
      have h3 : (Real.sqrt 3 : ℂ)^2=3 := by exact_mod_cast Real.sq_sqrt (show (0:ℝ)≤3 by norm_num)
      have h131 : (Real.sqrt (131 : ℝ) : ℂ)^2=131 := by exact_mod_cast Real.sq_sqrt (show (0:ℝ)≤131 by norm_num)
      have hn2 : (Real.sqrt 2 : ℂ) ≠ 0 := by exact_mod_cast (ne_of_gt (Real.sqrt_pos.2 (by norm_num : (0:ℝ)<2)))
      have hn3 : (Real.sqrt 3 : ℂ) ≠ 0 := by exact_mod_cast (ne_of_gt (Real.sqrt_pos.2 (by norm_num : (0:ℝ)<3)))
  2. For all indices $a, b, c, d \in \text{Fin } 3$, the identity $\rho_{(a, b), (c, d)} = \rho_{(c, b), (a, d)}$ holds for the density matrix $\rho = \text{ppt\_qutrit\_bell\_state}$. automation

    After introducing the four indices, `fin_cases` exhaustively breaks the goal into all $3^4 = 81$ possible index tuples. For each case, `simp` unfolds the definitions of `ppt_qutrit_bell_state` and `ppt_qutrit_bell_vector`, expands the sum of the four Bell projectors using `Fin.sum_univ_four`, and evaluates the matrix and outer product applications (`Matrix.sum_apply`, `Matrix.smul_apply`, `Matrix.vecMulVec_apply`, `Complex.real_smul`, `map_ofNat`). The tactic `field_simp` clears denominators using the non-zeroness of $\sqrt{2}$ and $\sqrt{3}$, `ring_nf` normalizes the resulting polynomial expressions, the square-root identities are substituted, and `norm_num` confirms the numerical arithmetic evaluates to 0.

    lines 15–19

    uses Fin.sum_univ_four, Matrix.sum_apply, Matrix.smul_apply, Matrix.vecMulVec_apply, Complex.real_smul, map_ofNat

      intro a b c d
      fin_cases a <;> fin_cases b <;> fin_cases c <;> fin_cases d <;>
        simp [ppt_qutrit_bell_state, ppt_qutrit_bell_vector, Matrix.sum_apply, Matrix.smul_apply,
          Matrix.vecMulVec_apply, Fin.sum_univ_four, Complex.real_smul, map_ofNat] <;>
        field_simp [hn2,hn3] <;> ring_nf <;> simp [h2,h3,h131] <;> norm_num
  3. The rank-four two-qutrit Bell state is invariant under partial transposition on the first qutrit.

    The theorem statement is identical to the private lemma `in767_sourceptproof_source_partial_transpose` proved in the previous steps, so applying it concludes the proof immediately.

    line 24

    uses in767_sourceptproof_source_partial_transpose

      exact in767_sourceptproof_source_partial_transpose
Real.sq_sqrt
For any real number $x \ge 0$, the square of its square root is $x$: $(\sqrt{x})^2 = x$.
ne_of_gt
In a preorder, if $b < a$, then $a \ne b$.
Real.sqrt_pos.2
If a real number $x$ is strictly positive, its square root is also strictly positive.
Fin.sum_univ_four
A sum over the four elements of $\text{Fin } 4$ equals $f(0) + f(1) + f(2) + f(3)$.
Matrix.sum_apply
An entry $(i, j)$ of a sum of matrices equals the sum of the $(i, j)$-entries of each matrix.
Matrix.smul_apply
An entry $(i, j)$ of a scalar multiple of a matrix equals the scalar multiple of that entry.
Matrix.vecMulVec_apply
The $(i, j)$-th entry of the outer product matrix $\text{vecMulVec } w\ v$ equals $w(i) \cdot v(j)$.
Complex.real_smul
Scalar multiplication of a complex number by a real number equals multiplication by that real number cast to $\mathbb{C}$.
map_ofNat
A semiring homomorphism preserves numeral literals.
in767_sourceptproof_source_partial_transpose
For all $a, b, c, d \in \text{Fin } 3$, the $((a, b), (c, d))$-entry of `ppt_qutrit_bell_state` equals its $((c, b), (a, d))$-entry.

Written by a model (gemini-3.8-flash) from the Lean proof and the goal states Lean reported at each step. Lean verified the proof; this explanation is a reading aid and is not itself machine-checked.

Lean source view module on GitHub

open Matrixopen scoped BigOperators Matrix ComplexOrderset_option backward.isDefEq.respectTransparency falseset_option maxRecDepth 8192set_option maxHeartbeats 4000000set_option linter.unusedSimpArgs falseprivate lemma in767_sourceptproof_source_partial_transpose :    ∀ a b c d : Fin 3,      ppt_qutrit_bell_state (a,b) (c,d) = ppt_qutrit_bell_state (c,b) (a,d) := by  have h2 : (Real.sqrt 2 : ℂ)^2=2 := by exact_mod_cast Real.sq_sqrt (show (0:ℝ)≤2 by norm_num)  have h3 : (Real.sqrt 3 : ℂ)^2=3 := by exact_mod_cast Real.sq_sqrt (show (0:ℝ)≤3 by norm_num)  have h131 : (Real.sqrt (131 : ℝ) : ℂ)^2=131 := by exact_mod_cast Real.sq_sqrt (show (0:ℝ)≤131 by norm_num)  have hn2 : (Real.sqrt 2 : ℂ) ≠ 0 := by exact_mod_cast (ne_of_gt (Real.sqrt_pos.2 (by norm_num : (0:ℝ)<2)))  have hn3 : (Real.sqrt 3 : ℂ) ≠ 0 := by exact_mod_cast (ne_of_gt (Real.sqrt_pos.2 (by norm_num : (0:ℝ)<3)))  intro a b c d  fin_cases a <;> fin_cases b <;> fin_cases c <;> fin_cases d <;>    simp [ppt_qutrit_bell_state, ppt_qutrit_bell_vector, Matrix.sum_apply, Matrix.smul_apply,      Matrix.vecMulVec_apply, Fin.sum_univ_four, Complex.real_smul, map_ofNat] <;>    field_simp [hn2,hn3] <;> ring_nf <;> simp [h2,h3,h131] <;> norm_numtheorem ppt_qutrit_bell_state_partial_transpose_invariant :  ∀ a b c d : Fin 3,    ppt_qutrit_bell_state (a,b) (c,d) = ppt_qutrit_bell_state (c,b) (a,d) := by  exact in767_sourceptproof_source_partial_transpose
See it in the graphDiscuss this resultChallenge it