Bulletin of the American Physical Society
APS March Meeting 2022
Volume 67, Number 3
Monday–Friday, March 14–18, 2022; Chicago
Session Y40: Noise Reduction and Error Mitigation in Quantum Computing IIIFocus Recordings Available
|
Hide Abstracts |
Sponsoring Units: DQI Chair: Abhinav Kandala, IBM Room: McCormick Place W-196B |
Friday, March 18, 2022 8:00AM - 8:12AM |
Y40.00001: Revealing the role of photon-assisted quasiparticle tunneling and superconducting gap differences in transmon qubits Spencer Diamond, Valla Fatemi, Max Hays, Kyle Serniak, Pavel Kurilovich, Thomas Connolly, Heekun Nho, Leonid Glazman, Michel H Devoret Tunneling of nonequilibrium quasiparticles (QPs) is a source of decoherence in transmon qubits. This tunneling may occur either by pre-existing QPs tunneling across the Josephson junction or with the generation of QPs at the junction due to a photon-assisted tunneling process. In order to distinguish the contribution of photon-assisted QP tunneling to the overall QP tunneling rate, we measure the QP tunneling rate in a flux-tunable charge-parity-sensitive transmon. We observe a peak in the flux-dependence of the QP tunneling rate which we attribute to a difference in the superconducting gaps of the two aluminum films comprising the device. By examining qubit-state dependence of this peak, we infer that QPs rapidly relax in energy to a cold distribution with excess number in the lower gap film. With a model that accounts for gap difference and photon-assisted tunneling and generation, we determine that the photon-assisted tunneling process is responsible for approximately half of all QP tunneling. Furthermore, by enhancing the rate of the photon-assisted tunneling process with a thermal photon source, we show that photon-assisted generation at the Josephson junction accounts for approximately half the QP density in the device. |
Friday, March 18, 2022 8:12AM - 8:24AM |
Y40.00002: Controlled mitigation of quasiparticle tunneling in a 3D transmon Thomas Connolly, Pavel Kurilovich, Valla Fatemi, Spencer Diamond, Max Hays, Michel H Devoret Non-equilibrium quasiparticle tunneling determines an upper bound for the coherence time of superconductor-based qubits. An important source of quasiparticles and quasiparticle tunneling is photons with energy above twice the superconducting gap. By using a charge-parity-sensitive 3D transmon qubit, we directly measure the quasiparticle tunneling rate, and we evaluate non-ambiguously the effectiveness of microwave line filters and radiation shields for reducing the incident flux of high-frequency photons on the transmon. We find that its quasiparticle tunneling rate can be reduced below 1/(100 ms) with a combination of filtering and shielding measures. |
Friday, March 18, 2022 8:24AM - 8:36AM |
Y40.00003: Suppression of Phonon-mediated Quasiparticle Poisoning by Silicon Micromachining Matthew Snyder, Sohair Abdullah, Chuan-Hong Liu, David C Harrison, Shravan Patel, Chris D Wilen, Owen Rafferty, Spencer Weeden, Gabriel Spahn, Andrew L Ballard, Vito M Iaia, Jaseung Ku, Britton L Plourde, Robert McDermott Phonon-mediated quasiparticle poisoning associated with particle impacts results in correlated relaxation errors in superconducting qubits that share a common substrate. We describe approaches to suppress these correlated errors based on substrate modification by micromachining. In one approach, we use deep reactive ion etching (deep RIE) from the backside of the substrate to incorporate an array of scattering centers that prevent ballistic phonon propagation. In a separate approach, we use deep RIE to define phonon bottlenecks or moats that decouple individual qubits acoustically from neighboring devices. We use direct quasiparticle injection to characterize the suppression of quasiparticle poisoning in devices incorporating these mitigation approaches, and we compare the rate of correlated errors in these devices to that seen in baseline devices with no mitigation. |
Friday, March 18, 2022 8:36AM - 8:48AM |
Y40.00004: Simulating Entangled States and Purification Circuits Faster than the Stabilizer Tableaux Formalism Shu Ge, Stefan Krastanov Entanglement purification is crucial to the creation of reliable quantum networks. Network performance can be improved by optimizing the purification circuits for the particular error processes present on the network links and node hardware. Over the last few years, such improvements have been demonstrated by many entanglement circuit optimization works. In our research, we present how such numerical simulations and optimizations can be made much more efficient by employing a simpler and smaller representation for entangled states and the circuits acting on them. It is known that Clifford circuits are efficient to simulate on classical computers and sufficient to represent purification circuits. We restrict ourselves to an even smaller set of circuits that can be represented as permutations of the Bell basis states, thus enabling even more efficient entanglement purification simulations. We demonstrate the performance of this method in some optimization and machine learning design algorithms, creating state-of-the-art purification circuits. |
Friday, March 18, 2022 8:48AM - 9:00AM |
Y40.00005: High-fidelity dissipative stabilization of multipartite entanglement Emery Doucet, Leonardo Ranzani, Luke Govia, Archana Kamal Quantum reservoir engineering allows autonomous preparation and stabilization of entangled states in a quantum system, where carefully chosen always-on interactions with an auxiliary bath steer the system to the desired state. States prepared this way are stable over indefinitely long time intervals and robust to decoherence, in contrast to states prepared using typical gate-based methods. To date, dissipative stabilization of few-qubit Bell, W, and GHZ states has been demonstrated in a variety of quantum information processing platforms. Recent works have highlighted the importance of “exact” stabilization [1,2], which we expect to become increasingly important when preparing larger and more complex states with high fidelity. In this talk, We will introduce a platform-agnostic family of exact stabilization protocols that prepare entangled states of arbitrarily many qubits in a resource-efficient manner. We will motivate and discuss the design rules underlying the protocol, and show how the stabilization performance scales as the number of qubits is increased. |
Friday, March 18, 2022 9:00AM - 9:12AM |
Y40.00006: On-Demand Cavity Cooling and Reset with a Modular Dissipator Darian M Hartsell, Revanth Kondaveti, Haimeng Zhang, Azarin Zarassi, Eli Levenson-Falk Spurious thermal photons in readout resonators are a leading source of dephasing in superconducting qubits. To remove these photons, one can introduce a lossy interaction to the environment. To preserve the coherence of a logical qubit, we introduce a lossy "dissipator" based on a transmon qubit, a flexible and modular component which can be integrated into any superconducting qubit circuit. Using a parametric drive to induce coupling between the cavity and dissipator, we remove thermal photons from the cavity and dissipate them to the environment, thereby suppressing noise seen by a logical qubit coupled to the same readout resonator. We discuss theoretical, numerical, and experimental results demonstrating that this driven dissipative cooling provides fast cavity reset and can cool the system below the bath temperature. |
Friday, March 18, 2022 9:12AM - 9:48AM |
Y40.00007: Fault-Tolerant Quantum Computing with Silicon Photonics Invited Speaker: Mercedes Gimeno-Segovia Quantum computers promise a new paradigm of computation where information is processed in a way that has no classical analogue. However, the known problems for which quantum computers offer a computational advantage require long gate sequences and large number of qubits, which means that effective methods of noise mitigation and error correction must be at the core of the architectural design of any useful quantum computer. Photons make great qubits, they are cheap to produce, resilient to noise and the only known option for quantum networks. Most crucially, they can be efficiently manipulated with silicon photonics, an intrinsically scalable and manufacturable platform in which all the fundamental quantum gates can be implemented. In this talk, I will describe an architecture for universal fault-tolerant quantum computing supported by a silicon photonics platform. In particular, I will describe how its unique networking capabilities enable modular architectures with high thresholds. |
Friday, March 18, 2022 9:48AM - 10:00AM |
Y40.00008: ab initio computational searches for TLS in amorphous dielectrics Keith G Ray, Sebastien Hamel, Vincenzo Lordi, Yaniv J Rosen Superconducting qubits are subject to noise affecting resonance frequencies and loss due to interactions with two level systems (TLS). TLS have been determined to reside in dielectrics used in device fabrication and are most strongly coupled to the qubit when in dielectric volumes bathed in strong electric fields from the superconducting resonator. TLS, in abstract, could correspond to atomic rearrangements of the dielectric material that possess electric dipole moment changes and are separated by small energy differences, in order to couple to a roughly 5 GHz excitation. In order to identify and characterize TLS in amorphous dielectric materials, we performed computational searches for transition states utilizing forces generated from both density functional theory and classical potentials. We then analyzed the TLS by calculating the dipole moment changes and comparing the atomic displacements, energy barriers, and energy differences. |
Friday, March 18, 2022 10:00AM - 10:12AM |
Y40.00009: Understanding and Mitigating TLS-Induced High-Order Decoherence in Superconducting Qubits Ziwen Huang, Xinyuan You, Shaojiang Zhu, Anna Grassellino, Alexander Romanenko Two-level systems (TLSs) on uncontrolled device surfaces are believed to cause both depolarization and dephasing in superconducting qubits. They are related to a number of noise channels, including the dielectric, charge and flux noise. Understanding and mitigating the TLS-related decoherence have been a central task towards improving the performance of superconducting quantum processors. In this presentation, we focus on the high-order effect of qubit-TLS coupling on both qubit depolarization and dephasing. Based on these results, we introduce mitigation strategies to reduce the loss of qubits while maintaining their controllability. |
Friday, March 18, 2022 10:12AM - 10:24AM |
Y40.00010: Noise mitigation using spectator quantum systems and continuous feedforward Andrew Lingenfelter, Aashish Clerk There has been significant recent interest in using a set of so-called “spectator” qubits to detect correlated noise fluctuations and ultimately improve performance of a distinct set of “data” qubits (see e.g. [1-3]). Here we theoretically analyze a continuous-in-time version of this idea: use a spectator quantum system (i.e. a qubit or cavity) in a continuous measurement and feeforward scheme to continuously cancel the influence of correlated noise on the “data” system. We discuss fundamental limits placed on this approach by quantum measurement constraints, and also compare against autonomous spectator schemes based on reservoir engineering ideas. In addition to helping improve qubit performance, we discuss how analogous strategies could be used to improve force sensors. |
Friday, March 18, 2022 10:24AM - 10:36AM |
Y40.00011: Flattening Floquet spectra of periodically driven quantum systems Ian Mondragon, Ziwen Huang, Jens Koch Controlling the energy dispersion of quantum systems has many applications, ranging from enhancing dephasing times in qubit systems to realizing novel many-body phenomena. In this regard, obtaining a dispersionless spectrum is of special interest. So far, most conventional approaches have studied flat dispersion relations on a case-by-case basis. Here we will discuss how to use periodic drives to systematically flatten the quasi-energy spectrum of a given system. For the special case of systems with chiral symmetry, we show that exact flattening can be achieved by means of periodic kicks. We conclude with two case studies: first, we illustrate how the bulk energy bands of Majorana wires can be flattened without destroying their topology, opening the possibility for enhanced interaction and disorder effects in these systems; second, we find that spectral flattening can help reduce the dephasing rate of superconducting qubits due to charge or flux noise. |
Friday, March 18, 2022 10:36AM - 10:48AM |
Y40.00012: Calibrated microwave measurements of qubit drive line components at millikelvin temperature Slawomir Simbierowicz, Volodymyr Monarkha, Suren Singh, Nizar Messaoudi, Philip Krantz, Russell E Lake Systematic errors in qubit state-preparation arise due to non-idealities in the qubit control lines such as impedance mismatch. Using a databased short-open-load calibration technique at a temperature of 30 mK, we determine the impedance matching of individual qubit drive line components: microwave attenuators and coaxial cables. We report calibrated 1-port S-parameters and interpret the results using a master equation simulation of all XY gates performed on a single qubit. Considering only drive pulse distortion due to reflections as the non-ideality, the data and simulations establish the required impedance matching threshold to enable a single-qubit gate fidelity of over 99.9%. |
Friday, March 18, 2022 10:48AM - 11:00AM |
Y40.00013: Generalized fast quasi-adiabatic population transfer for improved qubit readout, shuttling, and noise mitigation. Felix Fehse, Marco David, Michel Pioro-Ladriere, Bill Coish Population-transfer schemes are commonly used to convert information robustly stored in some quantum system for manipulation and memory into more macroscopic degrees of freedom for measurement. These schemes may include, e.g., spin-to-charge conversion for spins in quantum dots, detuning of charge qubits between a noise-insensitive operating point and a measurement point, and parity-to-charge conversion schemes for qubits based on Majorana zero modes. A common strategy is to use a slow (adiabatic) conversion. However, in an adiabatic scheme, the adiabaticity conditions on the one hand, and accumulation of errors through dephasing, leakage, and energy relaxation processes on the other hand, limit the fidelity that can be achieved. We give explicit fast quasi-adiabatic (fast-QUAD) conversion strategies (pulse shapes) beyond the adiabatic approximation that allow for optimal state conversion. We account for a general source of classical Gaussian noise, and give analytic descriptions for the noise-induced transfer errors, allowing for noise-error mitigation strategies. We confirm the analytical results for generic models with numerical simulations. Further, we show that for a Pauli spin-blockade readout a fast-QUAD pulse should lead to a single-shot fidelity of better than 99.9 %, an improvement of approximately two orders of magnitude over what can be achieved with a t-linear detuning ramp under comparable conditions. |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700