arXiv:2610.22689
Circuit Synthesis for Variational Quantum Algorithms: The Depth of Algebra and Geometry
Boris Tsvelikhovskiy
Abstract
The analysis of Hamiltonian Variational Ansätze centers on three fundamental questions: which states are exactly reachable from a given initial state, what native control depth suffices for local and global synthesis, and whether a bounded circuit extension can expose a direction of decreasing cost at a suboptimal output. We address these challenges by leveraging the dynamical Lie algebras (DLAs) associated with the circuits and the geometry of the initial state's orbit under the action of the corresponding dynamical Lie group $G$. We show that an initial state has a closed projective orbit whenever it is a nondegenerate eigenstate of a control Hamiltonian. Combined with asymptotic QAOA convergence, this guarantees exact, finite depth ground state preparation for architectures employing diagonal problem Hamiltonians and irreducible stoquastic mixers. We bound the number of native control exponentials required for local unitary synthesis and state preparation using DLA commutator filtration growth alongside reachable orbit dimensions. Complementing these local results, we construct global frameworks on invariant sectors equipped with full unitary or special unitary dynamical Lie algebras. These constructions yield explicit applications to Pauli-generated systems and constrained maximum independent set multi-angle QAOA circuits; specifically, we show that their native control costs are directly determined by commutator filtration depth and independent-set toggle-group geometry, respectively. Finally, if the dynamical Lie group is compact, the observable $H_P$ generates an admissible evolution ($iH_P \in \mathfrak{g}$, where $\mathfrak{g}$ is the DLA), and the projective orbit of the initial state contains a ground state ray of $H_P$, then a bounded depth, identity-initialized circuit extension exhibits either first-order descent or negative curvature at every suboptimal circuit output.
AI Summary
The paper studies exact reachability, native-control synthesis depth, and descent directions in Hamiltonian variational circuits. It distinguishes the initial state’s dynamical-group orbit from its generally larger cyclic invariant subspace, and counts depth as the number of native Hamiltonian exponentials. A nondegenerate eigenstate of any control Hamiltonian has a closed projective orbit, even when the dynamical group is not closed. Combined with established asymptotic QAOA convergence, this implies exact ground-state preparation at some finite alternating depth for diagonal costs and irreducible stoquastic mixers. Separately, an observable whose generator belongs to the dynamical Lie algebra attains its energy infimum on every reachable orbit. For commutator-filtration increments $c_j$ and stabilization depth $\mathcal D$, local unitary synthesis requires at most $\sum_j(2j-1)c_j\leq(2\mathcal D-1)\dim\mathfrak g$ pulses; state-specific bounds replace algebra dimensions by tangent-orbit ranks. The construction uses conjugated native generators and inverse-function-theorem charts. Stabilization depth is bounded by $2n-1$ for multi-angle MaxCut on paths, cycles, and complete graphs. Global constructions combine elementary-flow synthesis with standard decompositions: fully controllable Pauli-generated systems admit $O(\mathcal D4^n)$ unitary and $O(\mathcal D2^n)$ state-transfer bounds; Flip-or-Stay MIS controls admit $O(D^3\Delta_\Gamma)$ and $O(D^2\Delta_\Gamma)$, respectively. For compact dynamical groups, admissible costs, and reachable ground states, an identity-initialized extension of bounded depth exposes a nonzero gradient or negative curvature at every suboptimal output. The results do not provide efficient angle-finding algorithms; exact QAOA reachability is nonconstructive, local-neighborhood sizes are unquantified, and depth-bound tightness, hardware constraints, and noise remain open.
the author used OpenAI ChatGPT (GPT-5.6 Sol and GPT-6 Astra) to assist in reviewing relevant literature, verifying the correctness, and refining the exposition.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.