arXiv subjects

Florentin Reiter

Publications and source records attributed to Florentin Reiter.

At least 19 recordsLinked to original sources

arXiv2026

Support Vector Machine with a Scalable Quantum Kernel

Quantum support vector machines are classification algorithms that rely on quantum-generated kernels. The fidelity quantum kernel commonly used in quantum support vector machines suffers from exponential concentration as system size increases, preventing an efficient scaling beyond fewqubit systems. We introduce the Hamming quantum kernel, a classical post-processing method that is based on the same measurement outcomes as the fidelity quantum kernel. However, it avoids the exponential concentration problem by using the full measurement statistics rather than a single fidelity value. We evaluate the approach on both classical data (MNIST) and synthetic data generated from quantum circuits, using systems ranging from 2 to 27 qubits. Throughout the simulations, the Hamming quantum kernel outperforms the fidelity quantum kernel whenever 15 or more qubits are used. Furthermore, for synthetic quantum data, our method consistently outperforms the classical Gaussian kernel. This demonstrates that the Hamming quantum kernel improves the expressivity and robustness at larger qubit scales without requiring any additional quantum ressources.

quant-ph↗

arXiv2026

Beyond Unitary Quantum Simulation: Open-System Approaches to Quantum Chemistry toward Quantum Advantage

Quantum simulation is widely regarded as one of the most promising routes to genuine quantum advantage, yet most existing approaches to quantum chemistry are formulated in terms of closed-system, unitary dynamics and ground-state preparation within the Born--Oppenheimer approximation. In this review, we discuss a broader perspective motivated by the observation that naturally occurring quantum systems are rarely isolated and often reach physically relevant states only through relaxation, decoherence, and thermalization. We first examine what is and is not known about exponential quantum advantage in chemistry, emphasizing that coherent Hamiltonian simulation provides the clearest formal case for speed-up, while many open questions remain for realistic problems. We then discuss how dissipation might ideally be integrated into quantum chemistry on a fault-tolerant quantum computer, using recent proposals for chemically motivated dynamical simulation as a guiding vision. More generally, we highlight the practical appeal of this approach to enhancing the robustness of quantum algorithms.

quant-ph↗

arXiv2026

Quantum Fanout Gates in Constant Depth via Resonance Engineering

We present a novel implementation of an n-qubit fanout gate using resonance engineering. Our proposed mechanism uses Jaynes-Cummings interactions between multiple qubits and a common harmonic oscillator to realize a fanout gate at the system-level. Our theoretical analysis establishes upper bounds on the gate error, demonstrating linear infidelity scaling in constant time -- a favorable trade-off compared to a conventional CNOT decomposition. To validate the performance of our scheme at large system sizes, we exploit permutation symmetry to reduce the simulation complexity from exponential to polynomial in the number of qubits, enabling simulation up to 100 qubits. The results of this numerical analysis are consistent with our theoretical findings and allow us to characterize the performance well. Our gate will enable faster stabilizer readouts and could provide polynomial speedups in many quantum algorithms.

quant-ph↗

arXiv2026

Quantum theory for phonon lasing and non-classical state generation in mixed-species and single trapped ions

In this article we present a comprehensive theoretical investigation of phonon lasing with mixed-species trapped ions, as demonstrated in [T. Behrle, Phys. Rev. Lett. 131 (2023)], employing both a semi-classical mean-field description and a full quantum theory. We derive an analytic expression for the second-order coherence function, confirming the experimental observation of the system's lasing behaviour above threshold. Building on the successful implementation of the two-ion lasing scheme, we propose a novel approach for achieving phonon lasing with a single trapped ion, offering significant experimental advantages and making the implementation of multiple phonon lasers within a single setup feasible. Furthermore, we explore lasing in a squeezed basis and in different regimes of the Lamb-Dicke approximation, highlighting the potential to produce non-classical states with promising applications in precision sensing. Our analysis of a sensing protocol based on squeezed states, using experimentally feasible parameters, shows a sensitivity enhancement of up to two orders of magnitude.

quant-ph↗

arXiv2026

Encoding Numerical Data for Generative Quantum Machine Learning

Generative quantum machine learning models are trained to deduce the probability distribution underlying a given dataset, and to produce new, synthetic samples from it. The majority of such models proposed in the literature, like the Quantum Circuit Born Machine (QCBM), fundamentally work on a binary level. Real-world data, however, is often numeric, requiring the models to translate between binary and continuous representations. We analyze how this transition influences the performance of quantum models and show that it requires the models to learn correlations that are solely an artifact of the way the data is encoded, and not related to the data itself. At the same time, structure of the original data, like continuity, can be obscured in the binary representation, hindering generalization. To mitigate these effects, we propose a strategy based on Gray-codes that can be implemented with essentially no overhead, conserves structures in the data, and avoids artificial correlations in situations in which the standard approach creates them. Considering datasets drawn from various low-dimensional probability distributions, we verify that, in most cases, QCBMs using the reflected Gray code learn faster and more accurately than those with standard binary code. This shows that, as complement to specifically tailored circuit ans\"atze or data pre-processing, binary encodings can be used to introduce inductive biases into generative machine learning models.

quant-ph↗

arXiv2026

Random layers for quantum optimal control with exponential expressivity

A long-standing problem in quantum optimal control is finding an optimal pulse structure that leads to an efficient exploration of the unitary space with a minimal number of optimization parameters. We solve this problem by constructing parametrized pulse sequences from random-amplitude pulses grouped in layers with one optimization parameter per layer. We show that, when increasing the number of pulses, the resulting random unitaries converge exponentially fast to the uniform Haar-random ensemble, thus providing for an efficient exploration of the unitary space. Grouping the pulses into layers allows for lowering the total number of optimization parameters. We focus on two random-layer (RALLY) methods: In RALLY$_\text{T}$, time durations of the layers are optimized while the pulse amplitudes are randomly chosen beforehand, possibly even from a few discrete values. RALLY$_\text{A}$ optimizes a joint scaling factor of the random pulse amplitudes in each layer. We numerically validate the two methods by applying them to problems of unitary synthesis, ground-state preparation and state transfer in different quantum systems. For all problems considered, both methods approach an information-theoretic lower bound on the number of optimization parameters and outperform other commonly used algorithms. In gradient-free optimization, the RALLY methods are orders of magnitude more accurate with fewer figure-of-merit evaluations. The RALLY methods are promising for advancing quantum machine learning and variational quantum algorithms.

quant-ph↗

arXiv2026

Stabilization of cat-state manifolds using nonlinear reservoir engineering

We introduce a novel reservoir engineering approach for stabilizing multi-component Schrödinger's cat manifolds. The fundamental principle of the method lies in the destructive interference at crossings of gain and loss Hamiltonian terms in the coupling of an oscillator to a zero-temperature auxiliary system, which are nonlinear with respect to the oscillator's energy. The nature of these gain and loss terms is found to determine the rotational symmetry, energy distributions, and degeneracy of the resulting stabilized manifolds. Considering these systems as bosonic error-correction codes, we analyze their properties with respect to a variety of errors, including both autonomous and passive error correction, where we find that our formalism gives straightforward insights into the nature of the correction. We give example implementations using the anharmonic laser-ion coupling of a trapped ion outside the Lamb-Dicke regime as well as nonlinear superconducting circuits. Beyond the dissipative stabilization of standard cat manifolds and novel rotation symmetric codes, we demonstrate that our formalism allows for the stabilization of bosonic codes linked to cat states through unitary transformations, such as quadrature-squeezed cats. Our work establishes a design approach for creating and utilizing codes using nonlinearity, providing access to novel quantum states and processes across a range of physical systems.

quant-ph↗

arXiv2025

Learning Reduced Representations for Quantum Classifiers

Data sets that are specified by a large number of features are currently outside the area of applicability for quantum machine learning algorithms. An immediate solution to this impasse is the application of dimensionality reduction methods before passing the data to the quantum algorithm. We investigate six conventional feature extraction algorithms and five autoencoder-based dimensionality reduction models to a particle physics data set with 67 features. The reduced representations generated by these models are then used to train a quantum support vector machine for solving a binary classification problem: whether a Higgs boson is produced in proton collisions at the LHC. We show that the autoencoder methods learn a better lower-dimensional representation of the data, with the method we design, the Sinkclass autoencoder, performing 40% better than the baseline. The methods developed here open up the applicability of quantum machine learning to a larger array of data sets. Moreover, we provide a recipe for effective dimensionality reduction in this context.

quant-ph↗

arXiv2025

Two-media laser threshold magnetometry: A magnetic-field-dependent laser threshold

Nitrogen-vacancy (NV) centers in diamond are a promising platform for high-precision magnetometry. In contrast to the use of spontaneous emission in a number of NV-magnetometers, laser threshold magnetometry (LTM) exploits stimulated emission of NV centers by placing an NV-doped diamond inside an optical cavity. The NV laser system is predicted to reach a high magnetic-field-dependent contrast and coherent signal strength, leading to an improved magnetic field sensitivity combined with a high linearity. Here, we consider a two-media setup where the cavity additionally includes a vertical external cavity surface emitting laser. This optically active material compensates cavity losses at \SI{750}{nm} while still allowing for magnetic-field-dependent effects from the NV-diamond. We demonstrate a magnetic-field-dependent laser threshold and investigate the effects of pump laser induced absorption of the diamond. The experimental data is supported by an analytical simulation based on a rate model. Furthermore, we derive a generalized formula to compute the shot-noise-limited magnetic field sensitivity in the regime of high contrast yielding 49.07(33) pT/$\sqrt{\text{Hz}}$ for the present setup. Simulations with an optimized NV-diamond suggest that values down to 4.9 ft/$\sqrt{\text{Hz}}$ are possible.

physics.optics↗

arXiv2025

Learning agent-based approach to the characterization of open quantum systems

Characterizing quantum processes is crucial for the execution of quantum algorithms on available quantum devices. A powerful framework for this purpose is the Quantum Model Learning Agent (QMLA) which characterizes a given system by learning its Hamiltonian via adaptive generations of informative experiments and their validation against simulated models. Identifying the incoherent noise of a quantum device in addition to its coherent interactions is, however, as essential. Precise knowledge of such imperfections of a quantum device allows to devise strategies to mitigate detrimental effects, for example via quantum error correction. We introduce the open Quantum Model Learning Agent (oQMLA) framework to account for Markovian noise through the Liouvillian formalism. By simultaneously learning the Hamiltonian and jump operators, oQMLA independently captures both the coherent and incoherent dynamics of a system. The added complexity of open systems necessitates advanced algorithmic strategies. Among these, we implement regularization to steer the algorithm towards plausible models and an unbiased metric to evaluate the quality of the results. We validate our implementation in simulated scenarios of increasing complexity, demonstrating its robustness to hardware-induced measurement errors and its ability to characterize systems using only local operations. Additionally, we develop a scheme to interface oQMLA with a publicly available superconducting quantum computer, showcasing its practical utility. These advancements represent a significant step toward improving the performance of quantum hardware and contribute to the broader goal of advancing quantum technologies and their applications.

quant-ph↗

arXiv2025

Laser-enhanced quantum sensing boosts sensitivity and dynamic range

Magnetometers based on nitrogen-vacancy (NV) centers in diamond have emerged as the most important solid-state quantum sensors. However, ensembles are limited in optical contrast to typically a few percent and high-sensitivity variants usually possess only a few $\mathrm{\upmu}$T dynamic range. Here, we demonstrate a laser threshold magnetometry-based NV system that avoids these limitations. By integrating the NV centers into a laser cavity and showing magnetic-field-dependent shifts of the laser threshold, we observe 100\,\% optical contrast, i.\,e., we are able to entirely switch off the laser system with the NV centers magnetic resonance. At the same time we achieve strong output signals up to 50\,mW. The system exhibits a photon-shot-noise-limited (PSNL) sensitivity of $<$400\,fT/$\sqrt{\textrm{Hz}}$ for all vector components, which we demonstrate to improve super-linearly with contrast. The ratio of the sensing-relevant parameters PSNL sensitivity and dynamic range, that can be traded at the cost of each other, marks an improvement factor of up to 590 over typical fluorescence-based readout and vapor cell sensors while also adding vector magnetometry capabilities. Such performance improvements provide a perspective for a highly sensitive magnetometer, which could be operated outside a magnetically-shielded room. This could bring a new generation of sensors for applications including magnetoencephalography, magnetic navigation, and magnetic anomaly detection.

quant-ph↗

arXiv2025

Scaling Quantum Algorithms via Dissipation: Avoiding Barren Plateaus

Variational quantum algorithms (VQAs) have enabled a wide range of applications on near-term quantum devices. However, their scalability is fundamentally limited by barren plateaus, where the probability of encountering large gradients vanishes exponentially with system size. In addition, noise induces barren plateaus, deterministically flattening the cost landscape. Dissipative quantum algorithms that leverage nonunitary dynamics to prepare quantum states via engineered cooling offer a complementary framework with remarkable robustness to noise. We demonstrate that dissipative quantum algorithms based on non-unital channels can avoid both unitary and noise-induced barren plateaus. Periodically resetting ancillary qubits actively extracts entropy from the system, maintaining gradient magnitudes and enabling scalable optimization. We provide analytic conditions ensuring they remain trainable even in the presence of noise. Numerical simulations confirm our predictions and illustrate scenarios where unitary algorithms fail but dissipative algorithms succeed. Our framework positions dissipative quantum algorithms as a scalable, noise-resilient alternative to traditional VQAs.

quant-ph↗

arXiv2025

Scalable dissipative quantum error correction for qubit codes

Dissipative quantum error correction (QEC) autonomously protects quantum information using engineered dissipation and offers a promising alternative to error correction via measurement and feedback. However, scalability remains a challenge, as correcting high-weight errors typically requires increasing dissipation rates and exponentially many correction operators. Here, we present a scalable dissipative QEC protocol for discrete-variable codes, correcting multi-qubit errors via a trickle-down mechanism that sequentially reduces errors weight. Our construction exploits redundancy in the Knill-Laflamme conditions to design correction operators that act on multiple error subspaces simultaneously, thereby reducing the overhead from exponential to polynomial in the number of required operators. We illustrate our approach with repetition codes under biased noise, showing a fourfold improvement in the exponential suppression factor at realistic physical error rates. Our approach connects autonomous QEC for discrete-variable codes with demonstrated dissipative protocols for bosonic codes and opens up new avenues for traditional measurement-feedback QEC and fault-tolerant quantum operations.

quant-ph↗

arXiv2025

Fieldoscopy at the Quantum Limit

We demonstrate a novel concept for measuring time-varying electric field transients of petahertz-scale photons down to a single-photon regime. We observe a clear transition from classical to quantum nature of light that agrees with our Monte Carlo model. We reach unprecedented yoctojoule-level sensitivity and a dynamic range exceeding 90 decibels. We utilize this capability to measure time-dependent intrapulse light coherence - a regime inaccessible to conventional, time-averaged spectroscopy. This opens new avenues for quantum information, cryptography, and quantum light-matter interactions on sub-cycle time scales with attosecond precision.

quant-ph↗

arXiv2025

Extrapolation method to optimize linear-ramp QAOA parameters: Evaluation of QAOA runtime scaling

The Quantum Approximate Optimization Algorithm (QAOA) has been suggested as a promising candidate for the solution of combinatorial optimization problems. Yet, whether - or under what conditions - it may offer an advantage compared to classical algorithms remains to be proven. Using the standard variational form of QAOA requires a high number of circuit parameters that have to be optimized at a sufficiently large depth, which constitutes a bottleneck for achieving a potential scaling advantage. The linear-ramp QAOA (LR-QAOA) has been proposed to address this issue, as it relies on only two parameters which have to be optimized. Based on this, we develop a method to estimate suitable values for those parameters through extrapolation, starting from smaller problem sizes (number of qubits) towards larger problem sizes. We apply this method to several use cases such as portfolio optimization, feature selection, clustering and weighted maxcut. From results obtained on a noiseless quantum emulator, we evaluate the quantum runtime scaling for finding the optimal solution and compare it with that of classical methods. In the case of portfolio optimization, we demonstrate superior scaling compared to the classical runtime for the problem sizes of up to $28$ qubits that we consider in this work.

quant-ph↗

arXiv2025

Variational quantum thermalizers based on weakly-symmetric nonunitary multi-qubit operations

We propose incorporating multi-qubit nonunitary operations in Variational Quantum Thermalizers (VQTs). VQTs are hybrid quantum-classical algorithms that generate the thermal (Gibbs) state of a given Hamiltonian, with applications in quantum algorithms and simulations. However, current algorithms struggle at intermediate temperatures, where the target state is nonpure but exhibits entanglement. We devise multi-qubit nonunitary operations that harness weak symmetries and thereby improve the performance of the algorithm. Utilizing dissipation engineering, we create these nonunitary multi-qubit operations without the need for measurements or additional qubits. To train the ansatz, we develop and benchmark novel methods for entropy estimation of quantum states, expanding the toolbox for quantum state characterization. We demonstrate that our approach can prepare thermal states of paradigmatic spin models at all temperatures. Our work thus creates new opportunities for simulating open quantum many-body systems.

quant-ph↗

arXiv2025

Characterization of coherent errors in gate layers with robustness to Pauli noise

Characterization of quantum devices generates insights into their sources of disturbances. State-of-the-art characterization protocols often focus on incoherent noise and eliminate coherent errors when using Pauli or Clifford twirling techniques. This approach biases the structure of the effective noise and adds a circuit and sampling overhead. We motivate the extension of an incoherent local Pauli noise model to coherent errors and present a practical characterization protocol for an arbitrary gate layer. Notably, the coherent noise estimation is robust to Pauli noise. We demonstrate our protocol on a superconducting hardware platform and identify the leading coherent errors. To verify the characterized noise structure, we mitigate its coherent and incoherent components using a gate-level coherent noise mitigation scheme in conjunction with probabilistic error cancellation. The proposed characterization procedure opens up possibilities for device calibration, hardware development, and improvement of error mitigation and correction techniques.

quant-ph↗

arXiv2025

Spin coherence in strongly coupled spin baths in quasi-two-dimensional layers

We investigate the spin-coherence decay of NV$^-$-spins interacting with the strongly-coupled bath of nitrogen defects in diamond layers. For thin diamond layers, we demonstrate that the spin-coherence times exceed those of bulk diamond, thus allowing to surpass the limit imposed by high defect concentrations in bulk. We show that the stretched-exponential parameter for the short-time spin-coherence decay is governed by the hyperfine interaction in the bath, thereby constraining random-noise models. We introduce a novel method based on the cluster-correlation expansion applied to strongly-interacting bath partitions. Our results facilitate material development for quantum-technology devices.

quant-ph↗