Noise reduction via optimal control in a light-matter quantum system

summary

The gist

Quantum noise reduction below the shot noise limit is a signature of light-matter quantum interaction The correct choice of pulse times can reduce the noise in the quadrature field mode well below…

In short

The research developed an optimal control protocol using Gaussian pulses to reduce quantum noise in a light-matter system described by the Jaynes-Cummings model. By optimizing pulse times, researchers achieved noise reductions exceeding 80%, providing an experimentally feasible method for generating non-trivial squeezing using common quantum control techniques.

Key concepts

Squeezed States
These are special quantum states where the uncertainty in one measurable property, like position or phase, is reduced below the standard limit. They are valuable resources for advanced technologies such as gravitational wave detection and quantum computing because they enhance measurement precision.
Jaynes-Cummings Model (JCM)
This model describes the fundamental interaction between a quantum field (like light) and a two-level system (TLS). It is crucial for understanding how the light and matter exchange energy, which is necessary to study transient squeezing effects in this type of quantum system.
Optimal Control Strategies
This involves finding the best sequence of control inputs, specifically Gaussian pulses applied to the TLS. The goal is to deliberately alter these inputs so that they maximize the desired effect—in this case, enhancing transient quantum squeezing in a way that surpasses simple methods.
Circuit QED
This is an ideal physical platform for testing the protocol. It uses superconducting circuits, like a fluxonium coupled to an LC resonator, to realize light-matter interactions. This setup is highly relevant because it represents a common and feasible technology for implementing quantum control experiments.

Terminology used across episodes

This episode discusses

The paper

Noise reduction via optimal control in a light-matter quantum system · Read on arXiv

Francisco Albarran-Arriagada, Guillermo Romero, *Enrique Solano, Juan Carlos Retamal

Departamento de FĿcsica, CEDENNA, Universidad de Santiago de Chile (USACH) · Kipu Quantum

DOI: 10.1103/PhysRevResearch.7.023124

Transcript

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

Kai: We're looking at the paper: "Noise reduction via optimal control in a light-matter quantum system".

Mira: Here's the gist of the paper. Quantum noise reduction below the shot noise limit is a signature of light-matter quantum interaction The correct choice of pulse times can reduce the noise in the quadrature field mode well below the shot noise, reaching reductions of over 80% As the Jaynes-Cummings model describes a pivotal light-matter quantum system, our approach for noise reduction provides an experimentally feasible protocol to produce a non-trivial amount of squeezing with current technology Squeezed states refer to quantum states that exhibit reduced uncertainty in one of the conjugate variables, such as position and momentum or phase and amplitude In quantum optics, squeezed states have garnered significant attention due to the application to gravitational wave detection Additionally, squeeze states are an important resources for several tasks in quantum technologies such as quantum teleportation

14, 15: , continuos variable quantum computing

16: , and quantum error correction

17, 18: The process of creating and manipulating these states, known as squeezed states engineering

19–25: , involves deliberately altering the quantum fluctuations to enhance precision measurements improving the performance of quantum devices Mechanism of Transient Squeezing in JCM A quantum field interacting with a two-level system (TLS), described by the Jaynes-Cummings model (JCM)

26–31: , can experience a squeezing process during transient evolution, but the extent of reduction is limited, regardless of the initial state

32, 33: The amount of transient squeezing can be enhanced through collective effects increasing the number of atoms

34: An alternative strategy for increasing transient noise reduction relies on applying a periodic sequence of kicks to the two-level system

35: Optimal Control Strategies The search for nonclassical states in this system continues to be an exciting problem

36, 37: In this work, we search for optimal Gaussian pulse sequences to control the TLS that enhance the quantum mode transient squeezing in the JCM Our results show a noise reduction larger than 80% for different pulse widths and initial states by optimizing the pulse time only This work provides a path to get non-trivial squeezing using only Gaussian control over a TLS, which is common in almost all quantum platforms, opening the door to larger degrees of squeezing by using a general control over the TLS in a JCM Comparison of Optimization Methods We explore four different minimization strategies: forward sequential strategy (FSS), iterative disorder strategy (IDS), gradient-based iterative disorder strategy (GBIDS), and gradient-free iterative disorder strategy (GF-IDS) The FSS finds only seven different pulses in the time windows gt ∈

0, 10: , while the other strategies can find 15 different pulses, allowing further reduction of quantum fluctuations IDS and IDS-GF find equivalent solutions and are better than the pulse train given by FSS Experimental Feasibility in Circuit QED Circuit QED implementation is an ideal platform for realizing our noise reduction protocol Our proposal considers a fluxonium capacitively coupled to an LC resonator and driven by a voltage source Vg, as shown in the equivalent circuit of Fig.

Kai: 7 Considering a fluxonium such as in Ref.

58: with transition frequency ω10 = 2π × 1.09 GHz, coherence times T1 ∼ 40−120µs and T2 ∼ 10−35µs, we estimate the light-matter coupling as g = ω10/200 ∼ 2π×5 MHz Our proposed noise reduction protocol with Gaussian pulses occurs within the time scale gt = 10, which implies a time of t ∼ 0.14µs to reach a reduction over 80% This time is well below the coherence times in the current experimental situation We have proposed a protocol based on optimal control to reduce the quantum fluctuation in the bosonic mode of a light-matter system Specifically, we found optimal Gaussian pulses train to reduce the quantum fluctuation more than the 88% We note that IDS and IDS-GF have almost the same performance, but in our simulation, IDS is faster due to that the time grid necessary to get the result is not large Finally, this work paves the way for the use of optimal control to design protocols to enhance quantum features useful in quantum technologies Numerical Results Summary For gσ = 0.05 and α = √20, all cases under study have a fluctuation reduction larger than the 80%, being in a case close to 90% The maximal reduction is reached for α = √10 with R% = 84.59% using the IDS protocol In contrast, for α = √20, we obtain less fluctuation reduction, reaching only R% = 21.31% using five Gaussian pulses The maximal reduction is reached after the 30 pulses with R% = 88.73% when gσ = 0.05 and α = √20 Implementation Details The Hamiltonian that describes the circuit can be obtained using standard circuit quantization techniques

59: The resulting quantum Rabi model Hamiltonian is H = ħω0/2σz + ħωa†a + iħgσy(a†−a) The coupling constant is defined as g = Cc 2Cr(CJ+Cg)q1ZrZ The protocol involves applying a train of Gaussian pulses to control the TLS, which is implemented using a Gaussian pulse centered at time ti with width σ, amplitude omega0 and central frequency ωp Key Findings on Pulse Times Table I shows the pulse times for different strategies obtained for gσ = 0.05 and α = √6 The optimal pulse times vary significantly across the different control strategies, demonstrating the sensitivity of the noise reduction to the control sequence For instance, t1 is 0.99 for FSS and 1.57664117 for IDS The results show that few pulses are needed to reach a maximal fluctuation reduction, as seen in the case of α = √20 with gσ = 0.05 Circuit Model Derivation The circuit Hamiltonian is obtained using the Legendre transformation H = 1/2 (q + CGVg)TC-1 (q + CGVg) + U(ϕF, ϕr) The circuit Hamiltonian can be approximated in the limit Cc ≪ Cr,Cg as H = q 2F 2(CJ + Cg)/2 + Cg/(CJ+Cg)VgqF + Cg/(CJ+Cg)Cc/Cr Vgqr + 1/ (CJ+Cg)Cc/Cr qFqr + q 2r 2Cr / 2 + U(ϕF, ϕr) The final Hamiltonian in second quantization is H = 4ECN 2F - Cg/(CJ+Cg) 2eVgNF + EL 2θ 2F - EJ cos(θF+2πf)z>Qubit + Interaction z Resonator Experimental Feasibility Assessment The proposed protocol occurs within the time scale gt = 10, which implies a time of t ∼ 0.14µs to reach a reduction over 80% This time is well below the coherence times in the current experimental situation The model considers a fluxonium capacitively coupled to an LC resonator and driven by a voltage source Vg, which is feasible within superconducting circuits technology The ratio Cc/Cr ≈ 10−2 can be safely neglected regarding undesired resonator displacement Acknowledgments We acknowledge financial support from Agencia Nacional de Investigacion y Desarrollo (ANID): Financiamiento Basal para Centros Científicos y Tecnológicos de Excelencia (Grant No.

Paper summary: Kai: Building on what we just discussed, this paper titled "Noise reduction via optimal control in a light-matter quantum system" sets out to use optimal quantum control over the two-level system to enhance the transient noise reduction in the harmonic mode of that Jaynes-Cummings model.

Mira: The thesis is that we propose using a sequence of Gaussian pulses in a given time window, and they claim this specific choice of pulse times can reduce the noise in the quadrature field mode well below shot noise, reaching reductions over eighty percent.

Lev: The core contribution is showing that this approach for noise reduction provides an experimentally feasible protocol to produce a non-trivial amount of squeezing with current technology.

Kai: It really hinges on finding those correct pulse times, and they found that by optimizing the pulse time only, you can reduce the noise in different cases by more than eighty percent.

Mira: They are exploring several minimization strategies for this control: forward sequential strategy, iterative disorder strategy, gradient-based iterative disorder strategy, and gradient-free iterative disorder strategy.

Lev: The paper explores these four different minimization strategies to see which one performs best in finding the right pulse sequence for controlling the two-level system.

Kai: We talked about how the FSS only finds seven pulses in a time window from zero to ten, but IDS and IDS-GF can find fifteen different pulses instead.

Mira: They found that IDS and IDS-GF find equivalent solutions, and these are better than the pulse train given by the FSS.

Lev: So they’re essentially proving that these iterative strategies offer a path to finding a better pulse sequence for controlling the two-level system compared to just following a simple forward search.

Kai: This means we're looking at how this optimization works in practice, and it's not just about getting an answer, but about finding the best possible way to control the TLS.

Conclusion: Mira: So wrapping up the discussion on "Noise reduction via optimal control in a light-matter quantum system" by Kai and Mira, the title suggests we're looking at how optimal control can tame noise in this specific quantum system.

Kai: It means that through the work of Albarran-Arriagada and colleagues, we’ve found a method using Gaussian pulses to reduce the noise in the quadrature field mode well below shot noise.

Lev: And for someone building hardware, it means we can design protocols that use this method within existing coherence times, which is what makes this result practical.

Mira: So, in simple terms, they are showing that you don't need to find the absolute best pulse sequence to get a significant noise reduction by optimizing just the timing.

Kai: It’s about using optimal control over the two-level system to enhance those transient squeezing features in a way that’s achievable with current technology.

More episodes

← Home