Numerically optimized amplitude-robust controlled-Z gate for ultracold neutral atoms with individual addressing capability

arXiv:2604.12279 · quant-ph · Submitted 2026-04-14 · Read on arXiv

Listen

Transcript

Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.

Kai: Today's paper: "Numerically optimized amplitude-robust controlled-Z gate for ultracold neutral atoms with individual addressing capability".

Mira: Numerically optimized amplitude-robust controlled-Z gate for ultracold neutral atoms with individual addressing capability addresses the challenge of achieving high-fidelity entangling gates in ultracold atomic arrays by developing a numerically optimized…

Kai: First, who's behind it and why it matters.

Title and authors: Kai: We started by looking at the title and authors of this paper, "Numerically optimized amplitude-robust controlled-Z gate for ultracold neutral atoms with individual addressing capability," and it immediately suggests a focus on both high-fidelity control and practical implementation in an atomic array setting.

Mira: The authors are from the Rzhanov Institute of Semiconductor Physics and Novosibirsk State University, which gives us a strong sense that this work is rooted in the intersection of precision physics and quantum information science.

Lev: As someone who deals with error correction, I'm interested in what kind of underlying physical assumptions these authors are making about the Rydberg blockade mechanism they are leveraging for this gate.

Kai: They are essentially proposing a new way to design the control pulses for a controlled-Z gate that is less sensitive to noise in the Rabi frequency, which is a huge practical concern when you're trying to make gates work reliably.

Mira: That focus on robustness against Rabi frequency variations suggests they are moving away from standard, simpler pulse designs toward something more finely tuned and analytically defined.

Lev: If this optimization holds up under the conditions of real hardware, it means we can potentially design a gate that is inherently less fragile when subjected to the inevitable fluctuations in laser intensity that occur during operation.

Kai: And the inclusion of "individual addressing capability" tells me they aren't just thinking about entangling pairs; they are thinking about how to target specific qubits within a larger, more complex atomic system.

Mira: That capability is key because it addresses the asymmetry introduced by things like thermal motion and imperfect beam alignment in experiments that use tightly focused lasers for excitation.

Lev: I'd be keen to know if the scheme they propose requires any exotic experimental setups beyond standard ultracold atom traps, or if it’s adaptable to more common setups.

Kai: They are demonstrating a numerical optimization scheme using tools like RydOpt and the Adam learning rate algorithm to find the best pulse parameters that maximize fidelity while incorporating penalties for variation and slope.

Mira: The use of these specific computational tools tells me they're tackling this problem as a complex optimization task, treating it as a high-dimensional search rather than just deriving an analytical solution.

Lev: From my side, I worry about the computational overhead; if the optimization becomes too complex to run quickly on a real experimental control loop, its theoretical advantage might not translate into practical speed.

Kai: They are showing that this numerical optimization can lead to pulse profiles like smooth time profiles defined by Bernstein basis polynomials, which are promising for experimental realization.

Mira: That is a concrete result; they're giving us specific pulse shapes that we can actually synthesize and test in the lab, which moves the paper out of pure theory.

The paper's summary: Kai: So, to summarize what this paper is really saying about "Numerically optimized amplitude-robust controlled-Z gate for ultracold neutral atoms with individual addressing capability," they are proposing a method that numerically tunes the laser pulse parameters to make the controlled-Z gate extremely robust against changes in the Rabi frequency.

Mira: They achieved this by using analytically defined phase profiles of the laser pulse, and their numerical optimization process incorporated penalties designed to make the gate performance insensitive to variations in both the absolute value and time variation of that Rabi frequency.

Lev: So, if I'm hearing this correctly, they are essentially finding a pulse shape that is optimized not just for a single operating point but for an entire range of possible operating points.

Kai: Exactly; they used an Adam optimizer to search through a pool of parameters to maximize the average fidelity while actively penalizing deviations in fidelity and the rate at which it changes with respect to the Rabi frequency.

Mira: That systematic search process is what allows them to achieve a level of amplitude robustness that they claim is almost an order of magnitude greater than prior attempts, specifically by optimizing against both variation and slope penalties.

Lev: That level of sensitivity suggests they are trying to create a gate that is hard to destabilize, which would be very beneficial for the stability required in error correction protocols.

Kai: Beyond just robustness against frequency drift, they also demonstrated how this protocol can handle variations in the individual Rabi frequencies for each of two atoms, which is crucial for their individual addressing goals.

Mira: That capability addresses that asymmetry where one atom might be driven differently than the other due to local experimental imperfections, which is a key hurdle when scaling up these systems.

Lev: If they can manage that level of control over individual atom excitation parameters simultaneously, it means the error budget for gate operations becomes much more manageable in a large-scale circuit.

Kai: In short, they're presenting a numerically optimized scheme for creating high-fidelity, individually addressable entangling gates by designing pulses that are inherently resilient to common experimental noise sources like Rabi frequency fluctuations.

The paper's improvements: Mira: The paper outlines the specific improvements they implemented to achieve this robustness, detailing the optimization steps used in Sec. II of the paper, where they describe how they reduced gate fidelity to variations in the Rabi frequency and analyzed sensitivity to pulse asymmetry and gradient of the Rabi frequency.

Lev: I'm interested in seeing exactly how those penalties work; does penalizing variation versus penalizing slope give them a better outcome than just using one or the other?

Kai: They started with an initial optimization to find a random pool of parameters, then they incorporated penalties for both fidelity variation and the slope of the fidelity dependence on Rabi frequency, aiming to make the gate performance "insensitive to the sign of the Rabi frequency variation."

Mira: By introducing that slope penalty, they are forcing the optimizer not just to find a local peak in fidelity but a region where the fidelity doesn't change rapidly as you move slightly away from that peak.

Lev: That suggests they are trying to create a flatter performance landscape around their target gate point, which is exactly what we need when implementing sequences of gates sequentially.

Kai: The further optimization involved increasing those penalties and expanding the range of Rabi frequencies used for the search, which resulted in a protocol robust to both variations in absolute Rabi frequency value and its time variation during the pulse.

Mira: Expanding the search space makes it clear they are not just looking for a lucky parameter set; they are building a solution that generalizes across a broader physical regime, which is much more valuable for experimental work.

Lev: If they managed to make it robust to both absolute value and time variation, that significantly simplifies the control challenge because we don't have to worry about the frequency drifting randomly mid-pulse.

Kai: They also compared their smooth twophoton amplitude-robust gate against the time-optimal and Levine-Pichler gates in two-photon excitation scenarios, showing that the smooth scheme strongly outperforms those with constant Rabi frequency in that configuration.

Mira: That comparison is telling because it shows that for multi-photon processes, a smoothly varying pulse is superior to one held at a fixed frequency, which directly supports the general theme of their amplitude robustness findings.

Conclusion: Kai: So, to wrap up on this paper, we've seen how this "Numerically optimized amplitude-robust controlled-Z gate for ultracold neutral atoms with individual addressing capability" is a sophisticated numerical tool for designing resilient Rydberg gates.

Mira: The main implication is that they have developed a method to design pulse profiles that are inherently robust against frequency fluctuations, leading to substantially more reliable entanglement operations in experimental setups.

Lev: For error correction, this suggests we have a design principle for creating gates that are less susceptible to the noise inherent in the control fields, which is a practical step forward.

Kai: Ultimately, this work provides specific pulse shapes and simulation results that can guide experimentalists on how to build more reliable quantum hardware with individual addressing capabilities.

Mira: It points toward using sophisticated numerical methods to shape laser pulses so they inherently handle the physics of amplitude variations better than previous static approaches.

Lev: I just hope that when we eventually see this realized in a lab, the performance measured will match these high theoretical estimates from their simulations without too much unexpected discrepancies.

Kai: We've discussed a lot about how this paper tackles robustness and individual addressing, and it seems like a solid piece of work for moving our experimental efforts forward.

Mira: Indeed, the focus on amplitude-robustness across multiple noise sources makes this a very compelling topic for condensed matter theorists to analyze further.

Lev: I think the next step will be figuring out how to integrate these findings into actual error correction circuits and see if they can provide a tangible benefit in terms of reduced gate error rates.

Kai: That sounds like where we need to go, exploring the practical implementation and testing of this optimized protocol.

Mira: It’s definitely a direction worth watching as researchers continue to push the boundaries on control schemes for neutral atom quantum computers.

Rzhanov Institute of Semiconductor Physics SB RAS · Novosibirsk State University · Institute of Laser Physics SB RAS

quant-ph

Submitted: 2026-04-14

Updated: 2026-10-01

Comments: 15 pages, 10 figures

Code: https://github.com/beterov/RobustCZ

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 80/100

The gist: Numerically optimized amplitude-robust controlled-Z gate for ultracold neutral atoms with individual addressing capability addresses the challenge of achieving high-fidelity entangling gates in

Key concepts

Amplitude Robustness: This refers to the gate's ability to maintain high fidelity even when the laser's Rabi frequency fluctuates. The optimization specifically targets making the gate performance insensitive not just to the absolute value of this frequency, but also to how it changes over time during the pulse, leading to much more stable operation.
RydOpt: This is a custom software package used for quantum optimization. It leverages JAX and Adam optimization algorithms to numerically find the best laser pulse parameters that maximize gate fidelity. It incorporates penalties that force the resulting protocol to be robust against frequency variations.
Individual Addressing Capability: The scheme is designed so that it works even when exciting two atoms separately with tightly focused lasers. This addresses real-world issues like thermal motion and beam instability by making the gate robust to differences in the individual Rabi frequencies used for each atom.
Rydberg Blockade Symmetric Controlled-Z (CZ) Gate: This is the core quantum operation being improved. It's a method to create an entangling gate where exciting one atom affects the probability of exciting another. The optimization focuses on making this specific type of two-atom interaction highly reliable across different experimental conditions.

Terminology

Fidelity — 44× in this paper · explained in 2 episodes
Robustness — 27× in this paper · explained in 5 episodes
Entanglement — 13× in this paper · explained in 4 episodes
Adam — 8× in this paper · explained in 2 episodes
Amplitude Robustness — 4× in this paper · explained in 1 episode
Alignment — 1× in this paper · explained in 1 episode
Computational overhead — 1× in this paper · explained in 2 episodes
Core — 1× in this paper · explained in 2 episodes
Phase Transitions — 1× in this paper · explained in 2 episodes

Summary

Numerically optimized amplitude-robust controlled-Z gate for ultracold neutral atoms with individual addressing capability addresses the challenge of achieving high-fidelity entangling gates in ultracold atomic arrays by developing a numerically optimized scheme that significantly increases robustness against variations in Rabi frequency and enables individual addressing.

The gist: The paper numerically optimizes a scheme for a neutral atom Rydberg blockade symmetric controlled-Z (CZ) gate to increase its robustness to variations in the Rabi frequency, demonstrating increased robustness almost by an order of magnitude compared to previous proposals.

Gate Protocol Optimization

The core of the work involves numerically optimizing a scheme for a neutral atom Rydberg blockade symmetric controlled-Z (CZ) gate using an amplitude-robust approach. This optimization utilizes a custom package called RydOpt, which is designed for the quantum optimization of Hamiltonian dynamics with analytically defined controls leveraging the JAX library and implementing the Adam optimization algorithm. The objective of this numerical process is to maximize gate fidelity by incorporating penalties for both the variation and slope of gate fidelities with respect to Rabi frequency variations.

Robustness Against Rabi Frequency Variations

The optimized protocol demonstrates amplitude robustness, that is an order of magnitude greater than in the previous proposal. This robustness was achieved by using analytically defined phase profiles of the laser pulse and optimizing them across a range of Rabi frequencies. The optimization steps involved:

Initial optimization using standard procedures to find a random pool of parameters.
Optimization incorporating penalties for both fidelity variation and the slope of the fidelity dependence on Rabi frequency, aiming to make the gate performance insensitive to the sign of the Rabi frequency variation.
Further optimization by increasing penalties and expanding the range of Rabi frequencies used for optimization, resulting in a protocol that is robust to both variations in absolute Rabi frequency value and its time variation during the pulse.

Application to Individual Addressing

The scheme is specifically designed for individual addressing capability when considering two atoms individually excited by tightly focused laser beams, which introduces asymmetry due to residual thermal motion and beam pointing instability. The protocol addresses this asymmetry by being robust not only to variations in the global Rabi frequency but also to variations in the individual Rabi frequencies for each of the two atoms, which can be introduced by varying one Rabi frequency relative to the other.

Performance in Two-Photon Schemes

The protocol is investigated for both single-photon and two-photon Rydberg excitation schemes. For two-photon excitation, where a Hamiltonian describes a system with detunings from intermediate states, the simulation compared the amplitude-robust gate against time-optimal and Levine-Pichler gates. The results showed that the smooth twophoton amplitude-robust gate strongly outperforms the gate protocol with constant Rabi frequency in this configuration. Furthermore, Monte Carlo simulations at finite temperatures demonstrated a clear advantage of our scheme for single-photon laser excitation.

Experimental Implementation Considerations

The numerical optimization yielded several types of pulse profiles suitable for experimental implementation:

Rectangular pulses with constant Rabi frequency and detuning.
Smooth time profiles defined by Bernstein basis polynomials, which offer substantial robustness improvement at the final optimization step compared to the constant Rabi frequency case.
Smooth pulse shapes with a constant Rabi frequency but finite rise and fall times, which are considered advantageous for experimental implementation despite having inferior robustness compared to the rectangular pulse case in certain metrics.

Finite Temperature Simulations

Monte-Carlo simulations were performed at finite temperatures of trapped atoms to assess performance under realistic conditions. For two-photon excitation using wavelengths 420 nm and 1013 nm, the simulation results showed that the amplitude-robust protocol substantially outperforms the time-optimal scheme when considering atomic motion. The main source of infidelity in this context was identified as the imbalance in Rabi frequencies for the first and second steps of laser excitation when the atoms are moving in both radial and axial directions. This imbalance leads to undesirable light shifts that strongly reduce gate performance, which can be partially mitigated by using Rydberg beams with waists slightly larger than the trap waist or by increasing potential depth.

Conclusion on Scheme Suitability

The paper concludes that the amplitude-robust gate protocol is perfectly suited for single-photon Rydberg excitation, which is commonly used in experiments with strontium atoms. For two-photon excitation in rubidium and cesium atoms through the second excited P state, it provides only a marginal advantage. However, it is advantageous for two-photon excitation schemes through the first excited state due to reduced difference in the required wavelengths and its ability to compensate for Doppler effects in counterpropagating beam geometry. The scheme is expected to further improve gate fidelity by overcoming imperfect matching of beam profiles in radial and axial directions when applied to three-photon Rydberg excitation.

Improvements for AI systems

Here are specific improvements that could be made to AI systems, derived from the scientific concepts presented in this paper:

I could improve the robustness of reinforcement learning (RL) agents or control policies in physical simulation environments by implementing an amplitude-robust gate protocol for state preparation or entanglement operations within those simulations.
The improved system could perform high-fidelity quantum operations (like controlled-Z gates) in near-real hardware environments where laser intensity variations (Rabi frequency noise) are present, achieving robustness orders of magnitude better than current methods.
I can improve the accuracy and efficiency of quantum simulation algorithms by incorporating analytically defined, amplitude-robust phase profiles into the control sequence used to evolve Hamiltonian dynamics, leading to significantly lower gate infidelity caused by frequency drift.
The system could be adapted for individual addressing capabilities in complex quantum architectures (like atom arrays), enabling targeted manipulation of specific qubits without disturbing neighbors, even under conditions of thermal motion and beam pointing instability.
The AI system can be trained to optimize control parameters (like pulse shapes or phase profiles) by incorporating gradient penalties related to the sensitivity of gate fidelity to variations in Rabi frequencies, effectively learning a robust control strategy rather than just a single-point optimum.
For two-photon quantum processes, I can develop models that account for the asymmetry between different laser excitation schemes (e.g., 5P vs 6P states), allowing the AI to select the optimal excitation path based on experimental conditions (like wavelength matching) to minimize infidelity caused by mismatched Rabi frequencies.
The system can be enhanced for high-temperature quantum computing scenarios by using Monte Carlo simulations integrated into its training loop, allowing it to predict and compensate for errors arising from thermal motion of atoms in optical traps.

These improvements would allow the AI system to perform more reliable, scalable, and robust quantum control tasks in physical hardware or complex simulation environments.

Abstract

We numerically optimized a scheme for a symmetric controlled-Z (CZ) gate based on the Rydberg blockade in neutral atoms, with the aim of increasing its robustness to variations in the Rabi frequency. The scheme exploits analytically defined phase profiles of the laser pulse and proves almost an order of magnitude more robust to Rabi-frequency variations than previously proposed protocols. We designed this amplitude-robust protocol with a smooth pulse shape and optimized it for a finite Rydberg blockade strength, which makes it suitable for experimental implementation in both single-photon and two-photon configurations. We further show that the protocol can be applied to individually addressed Rydberg excitation, accounting for the asymmetry of the Rabi frequencies of two atoms excited by tightly focused laser beams. This allows the effects of residual thermal motion of the trapped atoms to be reduced. Finally, we examined the performance of the protocol for single-photon and two-photon Rydberg excitation schemes at finite blockade strength, and demonstrated its advantages for individual addressing at finite temperatures of the trapped atoms.