arXiv:2609.40164
Spectral gaps and slow modes of Pauli rotations and random quantum circuits
Ziyuan Dong, Xiang Fan
Abstract
We determine which representation sectors control the slowest convergence of random Pauli rotations and several local random quantum circuits. For $d=2^n$, the continuous Pauli walk has exact all-representation gaps $$Δ_{\mathsf{SU}(d)}=\frac{d-8}{8(d-1)}\quad(n\ge4),\qquad Δ_{\mathsf{PU}(d)}=\frac{d(d-3)}{8(d^2-1)}\quad(n\ge3).$$ The special-unitary gap is attained in $\bigwedge^8\mathbb{C}^d$ and is invisible to balanced moments, while the projective gap is already attained in a balanced fourth-moment representation. We obtain exact discrete and orthogonal Pauli analogues, exact finite-Clifford chain/path gaps, and dimension-independent all-representation Haar brickwork lower bounds; on three qubits every weighted Haar-$\mathsf{SU}(4)$ edge walk reduces to one universal principal-angle constant. For real complete-graph models, the entire ordinary fourth-tensor gap reduces to a quadratic-size occupation operator, is simple for every $n\ge4$, and has an expansion beginning $$Δ_n^{(4)}=\frac{10}{9n}-\frac{88}{105n^2}+O(n^{-3}),$$ with competing branches separated only beyond all algebraic orders. For the two finite real Clifford complete-graph walks, this fourth-tensor mode is the exact all-representation group gap for every $n\ge144$; in the same range it simultaneously governs every ordinary even tensor degree $k\ge4$. Thus different representation scopes can have genuinely different slow modes.
AI Summary
The paper asks which representation sectors determine the slowest convergence of Pauli-rotation walks and local random circuits, distinguishing all-representation group gaps from fixed-moment gaps and tracking central phases and generated targets. For continuous Pauli rotations with $d=2^n$, it obtains exact gaps $\Delta_{\mathrm{SU}(d)}=(d-8)/[8(d-1)]$ for $n\ge4$ and $\Delta_{\mathrm{PU}(d)}=d(d-3)/[8(d^2-1)]$ for $n\ge3$. The former is attained in $\bigwedge^8\mathbb C^d$ and missed by balanced moments; the latter appears at balanced order four. It also determines discrete and orthogonal analogues, including first attainment of the orthogonal group gap at ordinary tensor degree eight. For projective Clifford brickwork on an open chain, the exact all-representation gap is $1-(4/5)\cos(\pi/n)$; uniform random-edge sampling divides this by $n-1$. Actual generated targets have small-size corrections. Representation-uniform Haar brickwork bounds are $>1/40$, $>1/135$, and $>1/3300$ for unitary and two real ensembles. Weighted three-qubit Haar gaps reduce to one unevaluated principal-angle constant. For real complete-graph fourth tensors, a quadratic-size occupation operator determines a simple gap, with $\Delta_n^{(4)}=10/(9n)-88/(105n^2)+O(n^{-3})$. Competing symmetry branches agree to all algebraic orders but have exponentially bounded splitting. For finite real Clifford walks, this mode determines the group gap and every even tensor-degree gap $k\ge4$ for $n\ge144$. Proofs combine projection comparisons, Pauli-label graph spectra, torus-weight divisibility, code-based witnesses, finite-group induced modules, and mean-field localization. Exact Haar group gaps, the three-site constant, smaller saturation thresholds, and leading branch-splitting asymptotics remain open.
OpenAI's ChatGPT (GPT-5.6 Sol and GPT-6 Astra) assisted with exploratory proof development, computational checks, literature searches, and English-language polishing
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- DOI
- 10.5281/zenodo.23099545
- Comments
- 79 pages