SQuInT 2021 Program
Full Program | Thursday | Friday | All Sessions | Posters | Talks
SESSION 3: Talks at Zoom | |
| 2:00pm-2:30pm | Shruti Puri , Yale University Scalable Fault-tolerant Quantum Computation with Kerr-Cats | Abstract With fault-tolerant QEC it is possible to solve a computational problem accurately by transforming a many-body quantum state with a sequence of faulty operations on imperfect quantum hardware. However, the enormous overhead cost of fault-tolerant protocols is the key bottleneck in practical realization of quantum computers. In this talk I will present the superconducting Kerr-cat qubit architecture. This qubit is realized in a parametrically driven non-linear oscillator and has some in-built protection against noise. Finally, I will discuss how this protected cat-qubit can significantly reduce the overheads for fault-tolerant QEC. |
| 2:30pm-2:50pm | Zoe Holmes, Los Alamos National Laboratory Barren plateaus preclude learning scramblers | Abstract Scrambling, the rapid spread of information through many-body quantum systems, is fundamental to a wide range of fields, from quantum chaos to thermalisation and black holes. However, given the complexity of many body quantum systems, scrambling can be hard to study using standard techniques. Recently, quantum machine learning (QML) has emerged as a promising paradigm for the study of complex physical processes. It is therefore natural to ask whether QML could be used to study scrambling. In this talk, we present a no-go theorem which restricts this possible use of QML. Specifically, we show that any QML approach used to learn the unitary dynamics implemented by a typical scrambler will exhibit a barren plateau, i.e. the cost gradient will vanish exponentially with the system size. As such, any QML algorithm to learn a scrambler will be untrainable. Crucially, in contrast to previously established barren plateau phenomena, which are a consequence of the ansatz structure and parameter initialization strategy, our barren plateaus holds for any choice of ansatz and initialization. Thus, previously proposed strategies for avoiding barren plateaus do not work here. More generally, given the close connection between scrambling and randomness, our no-go theorem also applies to learning random and pseudo-random unitaries. Consequently, our result implies that QML cannot be used to efficiently learn an unknown unitary process, placing a fundamental limit on QML. |
| 2:50pm-3:10pm | Jeremy Metzner, University of Oregon Using 'protected' modes in trapped ions to enable mid-algorithm measurements for CVQC* | Abstract Measurements of the motional states of trapped ions require coupling the motion to the ions’ internal spin states. These measurements, however, require detection of spin-dependent fluorescence. Photon scattering, giving rise to fluorescence, causes the ion to recoil, which generally decoheres the ions’ motional modes. This decoherence prevents mid-algorithm measurements, which are necessary for processes that require classical feedback. Overcoming this challenge is likely necessary for the viability of practical continuous variable quantum computing (CVQC) in trapped ions. To address this issue, we are investigating the use of ‘protected’ modes within chains consisting of an odd number of ions, where the center ion has zero displacement (3(N-1)/2 protected modes with N ions). As a demonstration we use a dual-species three-ion chain linear (88Sr+ -40Ca+ - 88Sr+), which enables us to simply address the center ion with global laser fields. We perform measurements of the heating rate and coherence time, via Ramsey interferometry, of these protected modes, to determine how much the decohering effects of photon scattering are suppressed. We are also developing models to minimize the effects of symmetry breaking of the chain due to radiation pressure, and non-linear coupling between modes, on the coherence time of the protected modes. *This research was supported by the U.S. Army Research Office through grant W911NF-19-1-0481. |
| 3:10pm-3:30pm | Karthik Chinni, University of New Mexico CQuIC Reliability of simulation on NISQ-era devices | Abstract The goal of NISQ-era quantum processors is to outperform classical computers on targeted tasks, such as quantum simulation, with moderately sized devices that are well controlled, but lack full fault-tolerant error correction. Here, we consider two different types of errors that could affect the reliability of NISQ devices: a background perturbation that leads to quantum chaos and Trotterization of the unitary dynamical map. In the first case, we consider simulation of the LMG model, which is integrable in the mean-field limit, but becomes chaotic in the presence of a background time-dependent perturbation. Here, we show that the quantities that depend on the global structure of the phase space, such as critical point estimates of the quantum phase transition, are robust to the presence of this perturbation while other aspects of the system such as the mean magnetization that depend on the local trajectories are fragile and cannot be reliably extracted from the simulator. Next, we analyze the effects of Trotterization on the simulation of p-spin models and identify the existence of “dynamical instability regions” in the Trotterized unitary map that are absent in the time evolution operator of the ideal p-spin model. We show that, even in the absence of chaos, Trotter errors proliferate in these dynamical instability regions, as the effective Hamiltonian associated with the Trotterized unitary becomes very different from the target p-spin Hamiltonian. |
| 3:30pm-3:50pm | Bibek Pokharel, University of Southern California Dynamically Generated Decoherence-Free Subspaces on Superconducting Qubits | Abstract Decoherence-free subspaces/noiseless subsystems (DFS/NS) preserve quantum information by identifying subspaces/subsystems of the Hilbert space that remain unaffected by decoherence. Identifying DFS/NS codes under collective decoherence is well-understood, and the resultant codes support scalable and universal quantum computation. While most experimental systems, including superconducting qubit-based devices, do not decohere collectively, it is possible to engineer the conditions for collective decoherence using dynamical decoupling (DD) sequences. We report on the creation and verification of DD-assisted DFS/NS codes on quantum processors provided by the IBM Quantum Experience. We compare the performance of a DFS/NS encoded qubit with its unprotected counterpart. We show that qubit lifetime can be improved substantially using DD-assisted DFS/NS codes. Furthermore, we exploit gate set tomography to characterize logical error channels and estimate logical gate error rates for the DFS/NS encoding. When combined with an analysis of qubit lifetimes for multiple simultaneously encoded qubits, we obtain a comprehensive picture of DFS/NS feasibility and scalability on near-term quantum processors. |
- Home
- Program
- Guide for Gather.Town
- Instructions for Presenters
- Submit Your Abstract
- Code of Conduct
- Subscribe to the SQuInT Mailing List
- Past SQuInT Meetings
SQuInT Chief Organizer
Akimasa Miyake, Associate Professor
amiyake@unm.edu
SQuInT Co-Organizer
Brian Smith, Associate Professor
bjsmith@uoregon.edu
SQuInT Local Organizers
Philip Blocher, Postdoc
Pablo Poggi, Research Assistant Professor
Tzula Propp, Postdoc
Jun Takahashi, Postdoc
Cunlu Zhou, Postdoc
SQuInT Founder
Ivan Deutsch, Regents' Professor, CQuIC Director
ideutsch@unm.edu
