Cross-Cut Stabilizer Rank Rank Formula¶
Claim/Theorem¶
Let \(H\) be any full-row-rank binary matrix whose row space is the stabilizer space \(\mathcal S\), and let the qubit set be partitioned as \(L\sqcup R\). Write
for the corresponding column split. Then the intrinsic cross-cut stabilizer rank from [[cross-cut-stabilizer-rank.md]] satisfies
Equivalently, if \(C=\ker H\) is the classical code with parity-check matrix \(H\), then
so \(\chi_L(\mathcal S)\) is exactly the usual connectivity function of the column matroid of \(H\), or equivalently of the classical code \(C\).
Proof sketch:
- From [[cross-cut-stabilizer-rank.md]],
because \(\mathcal S_L\cap\mathcal S_R=\{0\}\). 2. The projection of \(\mathcal S\) onto the \(R\) coordinates has dimension \(\operatorname{rank}(H_R)\), so
Similarly,
Substituting gives the rank formula. 3. For the code expression, use the standard projection identity
and \(\dim(C)=|L|+|R|-\dim\mathcal S\).
This makes the current frontier more concrete: proving a linear lower bound on \(\chi_L(\mathcal S)\) is the same as proving linear code or matroid connectivity across balanced qubit partitions.
Dependencies¶
- [[cross-cut-stabilizer-rank.md]]
Conflicts/Gaps¶
- This node makes \(\chi_L(\mathcal S)\) operational, but it does not yet provide a lower bound for the parity-check matrices coming from Quantum Tanner codes or other expander-style QLDPC families.
- The equivalence is for a fixed stabilizer space \(\mathcal S\) viewed as a binary linear space on qubit coordinates. Turning this into a hardware-depth theorem still requires [[stabilizer-cut-rank-functional.md]].
- The current graph does not yet contain a theorem saying that Tanner expansion of a chosen presentation forces linear column-matroid connectivity of its parity-check matrix across balanced cuts.
Sources¶
10.48550/arXiv.2109.1459910.48550/arXiv.0805.2199