Polynomial-time local-unitary equivalence of graph states
- Fields
- Topics
Problem
Does a deterministic classical algorithm running in \(n^{O(1)}\) time decide local-unitary equivalence of arbitrary graph states on \(n\) labelled qubits? The input consists of two finite simple graphs \(G\) and \(H\) on the same labelled vertex set \([n]:=\{1,\ldots,n\}\), for any integer \(n\geq1\). For \(F\in\{G,H\}\), define its graph state by Eq. (1):
With \(U(2)\) denoting the group of single-qubit unitary matrices, the required decision predicate is Eq. (2), with vertex labels fixed:
Source
Claudet explicitly asks whether local-unitary equivalence of graph states can be decided in polynomial time in Chapter 8, Conclusion, p. 107 of his thesis (arXiv version 2) [Cla25]. The input convention here keeps the vertex labels fixed, as in the quasi-polynomial algorithm of Claudet and Perdrix [CP25].
Progress
Local-Clifford equivalence, denoted by \(\sim_{\mathrm{LC}}\) and obtained by restricting each \(U_v\) to a unitary that normalizes the single-qubit Pauli group, has a deterministic decision algorithm (Lemma 1 and the following discussion, p. 2) with running time
\begin{equation} T_{\mathrm{LC}}(n)=O(n^4). \tag{3} \end{equation}The algorithm with running time in Eq. (3) also constructs an equivalence transformation when one exists; this solves the restricted Clifford problem, not the general unitary problem. [VDD04]
Claudet and Perdrix give an exact deterministic algorithm (Theorem 30, Section 3.5, p. 59:15) for the general problem with running time
\begin{equation} T_{\mathrm{LU}}(n)=n^{\log_2 n+O(1)}. \tag{4} \end{equation}By Eq. (4), the finite-input problem is decidable, but the established bound is quasi-polynomial rather than polynomial; Claudet’s subsequent thesis explicitly identifies the polynomial-time question as unresolved. [CP25], [Cla25]
The 2026 circle-graph result (Theorem 1) establishes
\begin{equation} G\text{ is a circle graph} \quad\Longrightarrow\quad \left(|G\rangle\sim_{\mathrm{LU}}|H\rangle \Longleftrightarrow |G\rangle\sim_{\mathrm{LC}}|H\rangle\right), \tag{5} \end{equation}In Eq. (5), a circle graph is the intersection graph of chords of a circle and \(H\) is any graph on the same labelled vertices; consequently this subclass admits a polynomial-time decision algorithm. [HMNC26]
Comment
No polynomial-time algorithm for arbitrary graph states, or result ruling one out under a stated complexity assumption, was found in the public literature checked through 9 September 2026. The open issue is the complexity of an already decidable problem, not the existence of an exact decision procedure; the circle-graph theorem only resolves a special class. The status audit used public primary sources and later-work searches; it is not an exhaustive citation-index audit. The catalog’s minimum LU–LC counterexample problem (op_c37650bfb81dbfc6) asks for the smallest failure of Clifford equivalence, whereas this entry asks the complexity of deciding unrestricted unitary equivalence.
References
- [VDD04]
- M. Van den Nest, J. Dehaene, and B. De Moor, “Efficient Algorithm to Recognize the Local Clifford Equivalence of Graph States,” Physical Review A 70, 034302 (2004).DOIarXiv
- [CP25]
- N. Claudet and S. Perdrix, “Deciding Local Unitary Equivalence of Graph States in Quasi-Polynomial Time,” in 52nd International Colloquium on Automata, Languages, and Programming (ICALP 2025), 59:1–59:20 (2025).DOIarXiv