A Compass on the Quantum State Sphere: The Hopf Ansatz for Arbitrary Pure-State Optimization

The Hopf ansatz introduces a binary-tree circuit structure that serves as both a state preparation routine and a coordinate system for arbitrary pure-state optimization, making it an executable compass for navigating Hilbert space. The core contribution is providing an explicit inverse map from amplitudes to physical angles, a diagonal induced metric for local distance information, and preparable normalized coordinate tangents that enable efficient gradient access.

The Hopf Ansatz Structure

The Hopf ansatz is defined as a binary-tree circuit based on generalized Hopf coordinates that recursively splits a Euclidean radius into orthogonal sine–cosine branches The real version provides a dimension-matched magnitude chart, while the complex version adds one phase per leaf This structure yields an explicit inverse map, a diagonal pullback metric, and exactly preparable normalized coordinate tangents The real case uses rotation gates defined by Ry(θ) = cos θ − sin θ / sin θ cos θ and the complex case uses generalized single-qubit rotations RC(θa, θb, θc).

Geometry and Metric Properties

The Hopf chart exposes the optimization geometry of the full-state family in closed form, with a pullback metric that is analytic and diagonal in both real and complex ansatze In the real case, this reduces to the usual sphere metric, while in the complex case, it uses the ambient round-sphere metric. Theorem 1 shows that this diagonal metric is analytic and allows Riemannian descent directions to be formed without hardware estimation or inversion of a dense quantum geometric tensor The diagonal entries are given by a formula involving ancestor indices and trigonometric functions.

Gradient Access and Scaling

Coordinate tangents can be exactly synthesizable quantum states within the same circuit family, which allows for coherent magnitude-layer batching and one indexed batch for the complex phase block The gradient formula is given by ∂θiE(θ) = 2 q gi,i(θ) Re[⟨ei(θ) H ψ(θ)⟩]. The framework organizes compiled gradient-access circuit families by tree layer, where magnitude tangents are grouped by layer and complex phase tangents use a leaf-indexed construction Theorem 4 proves that every gradient-access configuration has a compiled execution cost of O(nN).

Optimizer Performance

In deterministic real-state benchmarks, metric-aware Hopf optimizers reach numerical-precision median gaps The clearest gains are seen in smaller VQE mean final gaps and stronger concentration of metrology-inspired traces at numerical precision. For the complex ansatz, the magnitude block yields a diagonal metric where phase directions are weighted by local squared amplitudes The framework applies to objectives whose derivatives can be written locally as transition moments of Hermitian chain-rule observables.

Metrological Applications

The Hopf gradient-access framework extends to metrological costs whose local differentials are expectation-value differentials of Hermitian observables For a pure-state QFI objective, the minimized cost is CQ(θ) = 1/FQ(θ) + η, which requires the same Hopf tangent-state machinery as VQE. For fixed-readout CFI objectives, the chain-rule observable OCRam (θ) is constructed using chain-rule coefficients and transition moments The compiled-setting scaling for metrological objectives is O(Kread n) Cread cfg = O(Kread log N) Cread cfg.

Conclusion

The main message is structural: when arbitrary-state freedom is part of the model, the circuit can carry its own coordinate map, metric, and differential probes The Hopf ansatz turns universal state preparation into a navigable circuit framework, a compass for moving on the quantum state sphere. The framework applies most directly when the search object is an arbitrary normalized quantum state vector and the quantity being optimized admits a usable local derivative formula.

The gist

The Hopf ansatz introduces a binary-tree circuit structure that serves as both a state preparation routine and a coordinate system for arbitrary pure-state optimization.

How it works

The Hopf ansatz is defined as a binary-tree circuit based on generalized Hopf coordinates that recursively splits a Euclidean radius into orthogonal sine–cosine branches The real version provides a dimension-matched magnitude chart, while the complex version adds one phase per leaf This structure yields an explicit inverse map, a diagonal pullback metric, and exactly preparable normalized coordinate tangents.

Geometry and Metric Properties

The Hopf chart exposes the optimization geometry of the full-state family in closed form, with a pullback metric that is analytic and diagonal in both real and complex ansatze In the real case, this reduces to the usual sphere metric, while in the complex case, it uses the ambient round-sphere metric. Theorem 1 shows that this diagonal metric is analytic and allows Riemannian descent directions to be formed without hardware estimation or inversion of a dense quantum geometric tensor

Conclusion

The main message is structural: when arbitrary-state freedom is part of the model, the circuit can carry its own coordinate map, metric, and differential probes The Hopf ansatz turns universal state preparation into a navigable circuit framework, a compass for moving on the quantum state sphere.

1. Precision in State Optimization via Hopf Ansatz for VQE/Optimization Systems: The system can optimize arbitrary pure-state vectors by using a binary-tree circuit for arbitrary normalized real and complex state vectors, which acts as a compass for the search. This enables optimization beyond mere reachability to obtain coordinates, local distance information, and measurable directions essential for variational quantum algorithms like VQE.

2. Efficient Gradient Access via Layerwise Batching: The system can organize gradient access by organizing the compiled settings by magnitude layer and leaf-indexed phase family, which reduces the number of distinct compiled gradient-access circuit families logarithmically with Hilbert-space dimension. This allows for one indexed batch for the complex phase block and magnitude layer batching, meaning experimental repetitions contribute directly to one gradient component without postselection.

3. Geometry-Native Optimizer Implementation: The system can implement optimizer updates using the analytic diagonal metric, which allows Riemannian or state-sphere descent directions to be formed without hardware estimation or inversion of a dense quantum geometric tensor. Specifically, it uses the formula for the state-sphere gradient lift: G(θ):= gradS⋆ C = X i p i ci gi i(θ) ei = X i ci gi i(θ) ψ˙ i, where g is diagonal, no dense metric inversion is performed.

4. Metrology-Inspired Optimization: The system can optimize metrological costs like Quantum Fisher Information (QFI) or fixed-readout Classical Fisher Information (CFI). This is achieved by replacing the Hamiltonian with chain-rule observables, where the gradient formula becomes ∂θiC(θ) = 2 q gi,i(θ) Re[⟨ei(θ) OC(θ)ψ(θ)], allowing for optimization of probe states and sensing protocols.

5. Guaranteed Asymptotic Scalability: The system's compiled setting count is guaranteed to be logarithmic in the Hilbert-space dimension, specifically O(log N) Ccfg = O(log N) Ccfg(E) for VQE and metrology objectives. This means the number of required circuit families grows only logarithmically with dimension, regardless of the complexity of the objective or readout configuration (provided Kread is independent of N).