Southwest Quantum Information and TechnologyFifteenth Annual Meeting, February 21-23, 2013
| |
|
All Abstracts | Poster Abstracts | Talk Abstracts 1. Unitary Transformations in a Large Hilbert SpaceBrian Anderson, University of Arizona (Session 2 : Thursday from 3:15 - 3:45) Abstract. Quantum systems with Hilbert space dimension greater than two (qudits) provide an alternative to qubits as carriers of quantum information, and may prove advantageous for quantum information tasks if good laboratory tools for qudit manipulation and readout can be developed. We have implemented a protocol for arbitrary unitary transformations in the 16 dimensional hyperfine ground manifold of Cesium 133 atoms, using phase modulated rf and microwave magnetic fields to drive the atomic evolution. Our phase modulation waveforms are designed numerically using a variant of the highly efficient GRAPE algorithm. The fidelity of the resulting transformations is verified experimentally through randomized benchmarking, which indicates an average fidelity better than 97% across a sample of random unitaries. Our toolbox for quantum control is in principle applicable for a broad class of physical systems, such as large spins or anharmonic oscillators. 2. Squeezing of Spin Waves in Atomic EnsemblesBen Baragiola, Center for Quantum Information and Control, University of New Mexico (Session 10 : Saturday from 2:00 - 2:30) Abstract. Squeezing the collective spin of an atomic ensemble via QND measurement is based on the interaction between a cloud of atoms and a laser probe. When the shot noise resolution of the laser probe is below the projection noise fluctuations of the atoms, the resulting backaction can reduce the uncertainty for a collective atomic observable. Most current models of this process rely on idealized one-dimensional plane wave approximations of the underlying light-matter interaction, which are not appropriate for describing a real system consisting of a cigar-shaped cold atomic cloud in dipole trap interacting with a probe laser. We extend such models from first principles to include spatial dependence of both the light and of the atomic ensemble and apply it to QND spin squeezing for large-spin alkali atoms. The model includes spin waves, diffraction, paraxial modes, and optical pumping, derived by a full master equation description. We find that to optimally mode-match for spin squeezing one must consider not only collective scattering into the forward modes of the field, but also the effects of local decoherence from spatially-varying diffuse photon scattering. Surprisingly, in certain circumstances such local decoherence can be less destructive to spin squeezing than previously thought from phenomenological models. 3. Adaptive gate-set tomographyRobin Blume-Kohout, Sandia National Laboratories (Session 7a : Friday from 4:00 - 4:30) Abstract. Quantum information hardware needs to be characterized and calibrated. This is the job of quantum state and process tomography, but standard tomographic methods have an Achilles heel: to characterize an unknown process, they rely on a set of absolutely calibrated measurements. Many technologies (e.g., solid-state qubits) admit only a single native measurement basis, and other bases are measured using unitary control. So tomography becomes circular -- tomographic protocols are using gates to calibrate themselves! Gate-set tomography confronts this problem head-on and resolves it by treating gates relationally. We abandon all assumptions about what a given gate operation does, and characterize entire universal gate sets from the ground up using only the observed statistics of an [unknown] 2-outcome measurement after various strings of [unknown] gate operations. The accuracy and reliability of the resulting estimates depend critically on which gate strings are used, and benefits greatly from adaptivity. We demonstrate gate-set tomography and quantify the accuracy with which the individual gates can be estimated. 4. Experimental signatures of quantum annealingSergio Boixo, University of Southern California, Information Sciences Institute (Session 11 : Saturday from 3:30 - 4:00) Abstract. Quantum annealing is a general strategy for solving optimization problems with the aid of quantum adiabatic evolution. How effective is rapid decoherence in precluding quantum effects in a quantum annealing experiment, and will engineered quantum annealing devices effectively perform classical thermalization when coupled to a decohering thermal environment? Using the D-Wave machine, we report experimental results for a simple problem which takes advantage of the fact that for quantum annealing the measurement statistics are determined by the energy spectrum along the quantum evolution, while in classical thermalization they are determined by the spectrum of the final Hamiltonian only. We establish an experimental signature which is consistent with quantum annealing, and at the same time inconsistent with classical thermalization, in spite of a decoherence timescale which is orders of magnitude shorter than the adiabatic evolution time. For larger and more difficult problems, we compare the measurements statistics of the D-Wave machine to large-scale numerical simulations of simulated annealing and simulated quantum annealing, implemented through classical and quantum Monte Carlo simulations. For our test cases the statistics of the machine are - within calibration uncertainties - indistinguishable from a simulated quantum annealer with suitably chosen parameters, but significantly different from a classical annealer. 5. Quantum simulation and many-body physics with hundreds of trapped ionsJohn Bollinger, National Institute of Standards and Technology (Session 1 : Thursday from 8:30 - 9:15) Abstract. Many different quantum information protocols have been demonstrated with small linear chains of ions in rf (Paul) traps. I will describe our efforts to extend some of the techniques developed with small linear chains of ions to larger two-dimensional crystals of hundreds of ions formed in a Penning trap [1]. Our qubit (or spin) is the 124 GHz electron spin-flip transition in the ground state of Be+ in the 4.5 Tesla magnetic field of the trap. We control the spins with an effective transverse magnetic field obtained with 124 GHz microwaves [2]. By employing spin-dependent optical dipole forces, we engineer long-range Ising interactions (both ferromagnetic and anti-ferromagnetic) between the ion qubits [3]. We benchmark the interactions through measurements of mean-field spin precession [4]. I will describe the types of Ising interactions we can readily implement and discuss the prospects for simulating the transverse Ising model with hundreds of qubits. [1] T. Mitchell, J. J. Bollinger, D. Dubin, X. Huang, W. M. Itano, and R. Baughman, Science 282, 1290 (1998). [2] M. J. Biercuk, H. Uys, A. P. VanDevender, N. Shiga, W. M. Itano, and J. J. Bollinger, Quantum Information and Computation 9, 920 (2009). [3] K. Kim, M.-S. Chang, R. Islam, S. Korenblit, L.-M. Duan, and C. Monroe, Phys. Rev. Lett. 103, 120502 (2009). [4] J. W. Britton, B. C. Sawyer, A. C. Keith, C.-C. J. Wang, J. K. Freericks, H. Uys, M. J. Biercuk, and J. J. Bollinger, Nature 484, 489 (2012). 6. Instantaneous Quantum Circuits for Ising ModelsGavin Brennen, Macquarie University (Session 1 : Thursday from 10:15 - 11:00) Abstract. Statistical Mechanics has provided us with straightforward recipes to compute various physical quantities that can be experimentally probed on an interacting many-body system. But more often than not, the application of these recipes is computationally inefficient, as can be seen from very idealised systems. It may be expected that quantum algorithms could help in this regard. I will describe a scheme for measuring complex temperature partition functions of Ising models which, through appropriate Wick rotations, can be analytically continued to yield estimates for real ones. Notably, the kind of state preparations and measurements involved in this application can in principle be made "instantaneous", i.e. independent of the system size or the parameters being simulated. The estimation error is analysed numerically and analytically and shown to be compatible with prior art using larger depth quantum circuits. Also I'll describe some results on when the algorithm yields approximation scales with multiplicative rather than additive error which could have application in other contexts as well. Finally the dual problem concerning the BQP-hardness of computing partition functions for classical ferromagnetic and consistent Ising models in 2D a high but not perfect accuracy will be described. 7. Exploring adiabatic quantum computing trajectories via optimal controlConstantin Brif, Sandia National Laboratories (Session 6a : Friday from 2:00 - 2:30) Abstract. Adiabatic quantum computation (AQC) employs a slow change of the Hamiltonian, which helps keeping the system in the instantaneous ground state. When the evolution time is finite, dynamic trajectories corresponding to different forms of time-dependent control function(s) will result in different degrees of adiabaticity (quantified as the average ground state population during evolution). We employ optimal control methods to search for control functions that achieve the target final state while simultaneously maximizing the degree of adiabaticity. Exploring properties of optimal AQC trajectories in model systems elucidates dynamic mechanisms that minimize unwanted excitations from the ground state. 8. Magic state distillation with noisy Clifford gatesPeter Brooks, California Institute of Technology (Session 7b : Friday from 5:00 - 5:30) Abstract. A promising method for achieving universal fault-tolerant quantum computation is to supplement Clifford operations, which are sufficient for error correction but not a universal basis, with copies of certain single-qubit states called magic states. High-fidelity copies of these states can be prepared from noisy copies using state distillation protocols which use only Clifford gates. This process can proceed to arbitrarily high fidelity, assuming that the Clifford gates are perfect. In practice, imperfect Clifford operations will both reduce the efficiency of distillation and limit the achievable fidelity of the distilled state. This will be particularly relevant to quantum computation where the noise from Clifford operations is substantial, which will likely be the case with early demonstrations of fault-tolerant quantum computing. Recently, a number of interesting proposals have been made for more efficient state distillation protocols which use fewer ancillas to achieve a given error rate. We analyze and compare the efficiency and success probability for magic state distillation under these various proposals, taking into account the presence of imperfect Clifford operations. 9. Quantum Technology Taken to its LimitsTommaso Calarco, University of Ulm (Session 2 : Thursday from 3:45 - 4:30) Abstract. The full power of quantum coherence has not yet been tapped for everyday technological applications. The exquisite level of control of current atomic physics experiments may enable this, for instance in the field of quantum communication and quantum computing - but scalable quantum information processing requires extremely precise operations. Quantum optimal control theory allows to design the evolution of realistic systems in order to attain the best possible performance that is allowed by the laws of quantum mechanics. I will present a range of its applications to a variety of quantum technologies, and discuss its use in probing the ultimate limits to the speed, fidelity and size of the corresponding quantum processes. 10. Quantum Control: A Circuit-Based ClassificationCarlton Caves, University of New Mexico (Session 2 : Thursday from 2:15 - 2:45) Abstract. Control of the behavior of quantum systems, to make them do what we want them to do, instead of just what comes naturally, is fundamental to quantum information science. I will discuss a classification scheme that divides control and feedback techniques into three types: measurement-based control and feedback; coherent control and feedback; and quantum (noncommutative) control and feedback. The classification is based on how these techniques are represented in quantum-circuit diagrams and will be illustrated by examples. 11. Accurate quantum Z rotations with less magicChris Cesare, Center for Quantum Information and Control, Department of Physics and Astronomy, University of New Mexico (Session 7b : Friday from 4:30 - 5:00) Abstract. We present quantum protocols for executing arbitrarily accurate pi/2^k rotations of a qubit about its Z axis. Unlike reduced instruction set computing (RISC) protocols which use a two-step process of synthesizing high- fidelity "magic" states from which T = Z(pi/4) gates can be teleported and then compiling a sequence of adaptive stabilizer operations and T gates to approximate Z(pi/2^k), our complex instruction set computing (CISC) protocol distills magic states for the Z(pi/2^k) gates directly. Replacing this two-step process with a single step results in substantial reductions in the number of gates needed. The key to our construction is a family of shortened quantum Reed-Muller codes of length 2^(k+2)-1, whose distillation threshold shrinks with k but is greater than 0.85% for k <= 6. CC was supported in part by the Laboratory Directed Research and Development program at Sandia National Laboratories. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. 12. Cavity-enhanced non-demolition measurements for atom counting and spin squeezingZilong Chen, JILA and Department of Physics, University of Colorado at Boulder (Session 10 : Saturday from 2:30 - 3:00) Abstract. Cavity-enhanced probing of an atomic ensemble is an important tool for precision metrology. In particular, high resolution, non-destructive atom counting increases measurement or sensing bandwidth, and mitigates noise aliasing (Dick effect) in an atomic sensor. Achieving high resolution atom counting while preserving coherence can generate conditionally spin squeezed states that has phase sensitivity below the standard quantum limit. In recent years, there has been much interest in cavity-enhanced measurements for the above metrological applications. We consider fundamental measurement imprecision and scalings for cavity-enhanced measurements. As a particular example, we will discuss fundamental squeezing limits in Rb-87 for the clock and stretched hyperfine transitions, taking into account the multilevel structure. We will also discuss our experimental squeezing on the Rb-87 clock transition and give an outlook on current efforts to squeeze using the Rb-87 cycling transition. References: [1] Zilong Chen, Justin G. Bohnet, Joshua M. Weiner, Kevin C. Cox, and James K. Thompson, arXiv:1211.0723 [2] Zilong Chen, Justin G. Bohnet, Shannon R. Sankar, Jiayan Dai, and James K. Thompson, Phys. Rev. Lett. 106, 133601 (2011) 13. Is building a superconducting quantum computer actually feasible?Andrew Cleland, University of California - Santa Barbara (Session 4 : Friday from 8:30 - 9:15) Abstract. There has recently been tremendous progress in the performance of superconducting quantum circuits, especially in single qubit T1 and T2 coherence times, as well as in quantum measurement. Simple implementations of quantum algorithms have also been demonstrated. Are these advances sufficient to consider actually building a quantum computer? I will argue that the answer is (probably) affirmative, although such an effort would still be faced with many challenges, including how to achieve high-fidelity tune-up, control, and measurement of large numbers of qubits. A highly fault-tolerant approach is also needed; I will describe the surface code architecture, which provides what may be the most fault-tolerant scheme that includes topologically-protected logical operations. I will outline the basic principles and operation of this scheme, as well as prospects for the medium and long-term future of this area of research. 14. Graph Equitable Partitioning in Quantum Many-Body PhysicsDavid Feder, University of Calgary (Session 7c : Friday from 4:00 - 4:30) Abstract. The Hamiltonian for bosonic and fermionic particles hopping on lattices can be interpreted as the adjacency matrix of an undirected, weighted graph, usually with self-loops. The properties of these quantum many-body systems can therefore be analyzed in terms of graph theory. For example, the simple graph for non-interacting distinguishable particles is the Cartesian product of each particle's adjacency matrix; if these particles become indistinguishable, the graph 'collapses' via a graph equitable partition. Under various circumstances, equitable partitioning can allow for a more efficient determination of the eigenstates (and therefore the properties) of physically interesting quantum many-body systems. I will focus in particular on the ground states of the Bose and Fermi Hubbard models. 15. Minimax quantum tomography: the ultimate bounds on accuracyChris Ferrie, Center for Quantum Information and Control, University of New Mexico (Session 7a : Friday from 4:30 - 5:00) Abstract. There are many methods for quantum state tomography (e.g., linear inversion, maximum likelihood, Bayesian mean...). But none of them is clearly "the most accurate" for data of finite size N. Even the upper limits on accuracy are as yet unknown, which makes it difficult to say that a given method is "accurate enough". We address this problem here by (i) calculating the minimum achievable error for single-qubit tomography with N Pauli measurements, (ii) finding "minimax" estimators that achieve this bound, and (iii) comparing the performance of known estimators. 16. Adiabatic Quantum Computation with Neutral CesiumAaron Hankin, University of New Mexico (Session 11 : Saturday from 4:00 - 4:30) Abstract. We are implementing a new platform for adiabatic quantum computation (AQC) [1] based on trapped neutral atoms whose coupling is mediated by the dipole-dipole interactions of Rydberg states. Ground state cesium atoms are dressed by laser fields in a manner conditional on the Rydberg blockade mechanism [2,3], thereby providing the requisite entangling interactions. As a benchmark we study a Quadratic Unconstrained Binary Optimization (QUBO) problem whose solution is found in the ground state spin configuration of an Ising-like model. [1] E. Farhi, et al. Science 292, 472 (2000) [2] S. Rolston, et al. Phys. Rev. A, 82, 033412 (2010) [3] T. Keating, et al. arXiv:1209.4112 (2012) 17. Quantum many-body problems for identical particles: |