⚛️ lattice

Towards quantum computing Feynman diagrams in hybrid qubit-oscillator devices

This paper establishes a theoretical framework linking hybrid qubit-oscillator experiments to quantum field theory by interpreting characteristic function measurements as Feynman diagrams, and demonstrates through numerical simulations that these diagrams can be accurately reconstructed from Ramsey-type qubit measurements using maximum-likelihood estimation, offering a scalable bottom-up approach for probing lattice field theory simulators.

Original authors: S. Varona, S. Saner, O. Băzăvan, G. Araneda, G. Aarts, A. Bermudez

Published 2026-10-05

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: S. Varona, S. Saner, O. Băzăvan, G. Araneda, G. Aarts, A. Bermudez

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

In the vast landscape of modern physics, there is a powerful tool used to predict how particles interact and scatter: a method of drawing simple pictures that represent complex mathematical calculations. These pictures, known as Feynman diagrams, allow scientists to break down the chaotic behavior of quantum fields into manageable steps, much like a map helps a traveler navigate a dense forest. For decades, these diagrams have been the primary language for understanding the fundamental forces of nature, but they have always remained theoretical constructs, calculated on paper or by classical computers. Recently, a new frontier has opened where these abstract drawings are being tested against the reality of physical machines. This is the realm of quantum computing, where the goal is to use the strange rules of quantum mechanics to solve problems that are too difficult for any traditional computer. The challenge, however, is that these machines are still fragile and prone to errors, making it hard to know if the results they produce are truly correct. To solve this, physicists are looking for a way to verify their calculations by comparing them directly to the diagrams themselves, turning the theoretical map into a tangible object that can be measured.

A team of researchers has now taken a significant step in this direction by demonstrating how to reformulate the measurement of these Feynman diagrams as an estimation problem using a hybrid device that combines a single atom with a vibrating field. The experiment relies on a trapped ion, a single atom held in place by electromagnetic fields, which acts as a tiny probe. This ion is coupled to a quantum oscillator, a system that vibrates like a spring, representing the field of particles the scientists wish to study. By carefully controlling the interaction between the atom and the vibration, the researchers can encode information about the field's behavior into the atom's state. They then measure the atom to extract a specific mathematical signature called a characteristic function. This function contains a wealth of information about the quantum state of the vibration, including how it evolves over time and how it responds to external forces.

The breakthrough in this work is the realization that this characteristic function can be expanded into a series of terms that correspond exactly to the Feynman diagrams used in theoretical physics. Instead of just calculating these diagrams on a computer, the team showed that they can be reconstructed from measured data by treating the problem as a multi-parameter estimation task. They achieved this by using a technique called Ramsey interferometry, which involves preparing the atom in a specific state, letting it interact with the vibrating field for a set time, and then measuring the result. By repeating this process many times with different settings, they collected a large dataset of measurement outcomes. Using advanced statistical methods, they then worked backward from these outcomes to estimate the values of the individual terms in the expansion. In essence, they used the atom as a sensor to "read" the Feynman diagrams directly from the quantum system.

The researchers tested their method on two different types of quantum states: a standard squeezed state, which is a well-understood configuration where the uncertainty in the vibration is reduced in one direction, and a more complex non-Gaussian state, which involves a higher-order interaction that is much harder to simulate on a classical computer. For the standard state, they found that their estimates matched the theoretical predictions with high precision, confirming that their method works. For the more complex non-Gaussian state, where no exact mathematical solution exists, they performed numerical simulations to reconstruct the diagrams and verify that the results were consistent with the expected behavior. This is a crucial achievement because it provides a way to benchmark quantum computers: if a machine can accurately reproduce these diagrams, it is likely performing the complex calculations correctly.

To ensure their findings were robust, the team also simulated the experiment on a computer, taking into account the real-world imperfections that plague such devices. They modeled effects like the heating of the ion's motion, which introduces noise and can distort the results, as well as the decoherence of the atom itself. Their simulations showed that even with these imperfections, the method remains effective, provided the experiment is run in the right conditions. They identified specific ranges of parameters where the errors were minimized, allowing for the most accurate reconstruction of the diagrams. Furthermore, they explored how to handle the effects of temperature, showing that by using a specific mathematical framework known as the Schwinger-Keldysh formalism, they could account for thermal noise and still extract the correct information. This suggests that the technique is not limited to perfect, zero-temperature environments but can be applied to more realistic experimental setups.

The implications of this work extend beyond just verifying a single experiment. By successfully measuring these diagrams in a controlled setting, the researchers have laid the groundwork for a bottom-up approach to simulating complex quantum field theories. The idea is to start with a single oscillator and a single atom, as they did here, and gradually add more oscillators and atoms to simulate larger and more complex systems. This could eventually lead to the ability to study phenomena that are currently out of reach, such as the behavior of particles in the early universe or the properties of exotic materials. The ability to directly measure the components of these theories, rather than just the final result, offers a new way to validate quantum simulations and build confidence in their predictions.

In the end, this research bridges the gap between the abstract world of quantum field theory and the physical reality of quantum devices. It shows that the diagrams drawn on paper are not just mathematical conveniences but represent physical quantities that can be measured and verified. By turning the process of drawing a diagram into an experimental procedure, the team has opened a new path for exploring the quantum world. Their work suggests that in the near future, quantum computers could be used not just to solve problems, but to directly probe the fundamental structures of nature, providing a deeper understanding of how the universe works at its most basic level. The ability to estimate these diagrams with low error, even in the presence of noise, marks a significant step toward making quantum simulation a reliable tool for physics.

Technical Summary: Towards Quantum Computing Feynman Diagrams in Hybrid Qubit-Oscillator Devices

Problem Statement
The generating functional is a central object in Quantum Field Theory (QFT) encoding real-time dynamics and scattering information. While recent work has established that estimating the full generating functional for interacting scalar field theories belongs to the BQP complexity class (implying a potential quantum advantage over classical polynomial-time algorithms), verifying these results for arbitrary couplings remains difficult. However, in regimes where interactions are small, the generating functional admits an expansion in terms of Feynman diagrams. Directly estimating these diagrams from experimental data could serve as a rigorous benchmark for quantum computers simulating QFTs.

Current experimental realizations of hybrid quantum simulators (specifically trapped-ion devices) have successfully measured the characteristic function of bosonic oscillators, including non-Gaussian squeezed states. However, the connection between these interferometric measurements and the functional calculus of QFT—specifically the role of real-time propagators and Schwinger sources—has not been fully exploited to extract Feynman diagrams. This paper addresses the challenge of formulating the single-oscillator characteristic function as a functional path integral, expanding it into Feynman diagrams, and demonstrating how these diagrams can be reconstructed from qubit measurements using statistical inference.

Methodology
The authors develop a theoretical framework connecting hybrid qubit-oscillator interferometry to QFT path integrals and apply maximum-likelihood estimation techniques to reconstruct Feynman diagrams from simulated experimental data.

Functional Formulation:
- The characteristic function $\chi(\xi^*, \xi)$ of a bosonic state is promoted to a functional $\chi[J^*, J]$ by generalizing phase-space coordinates to time-dependent Schwinger source functions $J(t)$ and sinks $J^*(t)$ .
- This functional is identified as a vacuum (or generalized state) persistence amplitude in a $(0+1)$ -dimensional QFT.
- The authors consider initial states prepared via generalized squeezing operators $S_n(\zeta)$ acting on the vacuum, where $n$ denotes the order of squeezing (e.g., $n=2$ for standard Gaussian squeezing, $n=3$ for non-Gaussian tri-squeezing).
- The interaction potential $V_\lambda(t)$ is modeled as an impulsive switching of vertices at initial ( $t_0$ ) and final ( $t_f$ ) times, allowing the system to evolve under a free Hamiltonian with external sources in between.
Diagrammatic Expansion:
- Using path integral techniques, the characteristic functional is expanded into a perturbative series of Feynman diagrams.
- For $n=2$ (Gaussian), the expansion involves standard propagators and vertices.
- For $n \geq 3$ (Non-Gaussian), the vertices emit or absorb bundles of $n$ quanta, leading to distinct diagrammatic structures with specific symmetry factors.
- The formalism is extended to finite temperatures using the Schwinger-Keldysh contour, introducing a matrix propagator that accounts for thermal excitations and holes, and distinguishing between forward and backward time branches.
Quantum Tomography and Estimation:
- The measurement protocol involves a Ramsey interferometry scheme where a probe qubit is coupled to the oscillator via a state-dependent force (acting as the Schwinger source).
- Projective measurements of the qubit in the $X$ and $Y$ bases yield the real and imaginary parts of the characteristic functional.
- The problem of reconstructing the diagrams is cast as a multi-parameter point estimation problem. A maximum-likelihood estimator (MLE) is constructed to infer the coefficients of the Feynman diagram expansion from a set of Ramsey measurements taken at various source strengths and squeezing parameters.
- The analysis includes a detailed error budget, distinguishing between statistical errors (shot noise) and systematic errors (truncation of the diagrammatic series and thermal noise).
Numerical Simulations:
- Realistic simulations of trapped-ion devices (specifically $^{88}\text{Sr}^+$ ) were performed using QuantumOptics.jl.
- Imperfections modeled include motional heating (finite temperature initial states and heating during evolution), imperfect state preparation, and qubit decoherence.
- Techniques such as zero-temperature extrapolation and model corrections for heating rates were tested to mitigate systematic biases.

Key Contributions and Results

Path Integral Connection: The paper establishes a clear analogy between the characteristic function measured in hybrid qubit-oscillator experiments and the generating functional of a $(0+1)$ -dimensional QFT. It demonstrates that the time-dependent coupling in these experiments effectively acts as Schwinger sources.
Diagrammatic Reconstruction: The authors derive explicit Feynman diagram expansions for both Gaussian ( $n=2$ ) and non-Gaussian ( $n=3$ ) squeezed states. They show that the coefficients of these diagrams can be uniquely identified with the parameters of the characteristic functional.
Statistical Inference Framework: A maximum-likelihood estimation protocol is proposed to extract the coefficients of the Feynman diagrams from qubit measurement data. The study identifies optimal regimes for measurement parameters (source amplitude $\xi$ $ξ$ and squeezing $r$ $r$ ) that minimize the combined systematic and statistical errors.
- For $n=2$ , the estimated coefficients match the exact analytical values within statistical error bars for sufficient measurement shots ( $N \sim 10^6$ ).
- For $n=3$ , the method successfully reconstructs the complex coefficients of the non-Gaussian expansion, despite the lack of a closed-form analytical solution.
Thermal and Heating Effects:
- The formalism is generalized to finite temperatures using the Schwinger-Keldysh contour. The authors show that thermal effects introduce specific scaling factors (e.g., $(1+2n_B)$ ) to the diagram coefficients.
- Numerical results demonstrate that while non-zero temperature introduces systematic bias, it can be mitigated via zero-noise extrapolation or by incorporating heating parameters directly into the likelihood model.
- The inclusion of motional heating rates ( $\dot{n} \approx 300$ quanta/s) in the model significantly improves the accuracy of coefficient estimation compared to models that ignore heating.

Significance and Claims
The paper claims to provide a "first step" toward using noisy intermediate-scale quantum (NISQ) devices, specifically hybrid qubit-oscillator systems, to probe real-time dynamics in QFTs. Its primary significance lies in:

Benchmarking: It offers a perturbative benchmark for quantum simulators. By reconstructing Feynman diagrams, experimental results can be validated against perturbative predictions, building confidence in the device's ability to handle regimes beyond perturbation theory.
Foundational Synthesis: It bridges two of Feynman's revolutionary ideas: the diagrammatic representation of particle interactions and the circuit-based simulation of quantum systems.
Scalability: The work outlines a "bottom-up" approach. By starting with a single oscillator and a single qubit, the framework suggests a pathway to scaling up to multiple oscillators (discretized lattices) and multiple qubits to simulate full lattice field theories.
Practical Feasibility: Through realistic simulations including experimental imperfections, the authors demonstrate that Feynman diagrams can be reconstructed with low systematic and statistical errors using current trapped-ion technology, provided appropriate error mitigation strategies (like zero-noise extrapolation or heating-aware modeling) are employed.

The authors remain modest regarding the scope, noting that this work focuses on a $(0+1)$ -dimensional version of the problem to avoid ultraviolet divergences and renormalization complexities inherent in higher-dimensional QFTs. They posit that the techniques developed here serve as a complement to other benchmarking tools (like Monte Carlo or tensor networks) and pave the way for exploring non-perturbative phenomena in future, more complex hybrid quantum simulators.

More like this