Bulletin of the American Physical Society
APS March Meeting 2019
Volume 64, Number 2
Monday–Friday, March 4–8, 2019; Boston, Massachusetts
Session Y34: The Physics of Computing and Computing with PhysicsInvited
|
Hide Abstracts |
Sponsoring Units: DCMP Chair: Enrique Solano Room: BCEC 205A |
Friday, March 8, 2019 11:15AM - 11:51AM |
Y34.00001: Digital-Analog Quantum Computing Invited Speaker: Mikel Sanz Purely digital classical computers are rather recent devices which are a consequence of the impressive technological development in miniaturization of electronics and microchips. However, until only few decades ago, digital computers had not sufficient computational power for most applications, so they usually employed analog parts for specific hard calculations. The situation nowadays in quantum computing is similar due to the small number of coherent controllable qubits allowed by current platforms. Even doubling the number of qubits yearly, a purely digital approach with error correction will not allow us to solve relevant problems in the following decades. Here, we will explore in detail a universal digital-analog approach employing the ubiquitous Ising Hamiltonian as the resource. We use this global Hamiltonian, together with local rotations, to generate an arbitrary unitary and we find efficient protocols, polynomial in the number of single-qubit rotations, to produce relevant families of Hamiltonian, such as arbitrary two-body Hamiltonians or, in general, k-body Hamiltonians. We introduce the concept of banged digital-analog quantum computing, in opposition to the stepwise, when single-qubit rotations are much faster than the natural time-scale of the Hamiltonian, which allows us to compute without switching on/off the global interaction. Employing natural models of errors, we compare the performance of digital against both stepwise and banged digital-analog protocols showing that, in general, digital-analog approaches perform better both in time and fidelity. Finally, we will also show that this emerging digital-analog approach can be applied not only to quantum simulations, but also to quantum algorithms. Indeed, we provide an efficient digital-analog description of the Quantum Fourier transform, comparing its performance against the pure digital approach in the presence of errors. |
Friday, March 8, 2019 11:51AM - 12:27PM |
Y34.00002: The physics and challenges of neuromorphic computing with spintronic nanooscillators Invited Speaker: Julie Grollier TBD |
Friday, March 8, 2019 12:27PM - 1:03PM |
Y34.00003: MemComputing: leveraging memory and physics to compute efficiently Invited Speaker: Massimiliano Di Ventra In this talk I will discuss how to employ memory (time non-locality), in a novel physics-based approach to computation: Memcomputing [1, 2, 3]. As examples, I will show the polynomial-time solution of prime factorization, the search version of the subset-sum problem [4], and approximations to the Max-SAT beyond the inapproximability gap [5], using polynomial resources and self-organizing logic gates, namely gates that self-organize to satisfy their logical proposition [4]. I will also show that these machines are described by a Witten-type topological field theory, and they compute via an instantonic phase, implying that they are robust against noise and disorder [6]. The digital memcomputing machines we propose can be efficiently simulated, are scalable and can be easily realized with available nanotechnology components. |
Friday, March 8, 2019 1:03PM - 1:39PM |
Y34.00004: Benchmarking NISQ-era quantum processors Invited Speaker: Jay Gambetta
|
Friday, March 8, 2019 1:39PM - 2:15PM |
Y34.00005: Accelerating Deep Neural Networks with Analog Memory Devices Invited Speaker: Charles Mackin Over the next few years, special-purpose hardware accelerators based on conventional digital-design techniques will optimize the GPU framework for Deep Neural Network (DNN) computations, increasing speed and reducing power for both “training” and “forward-inference.” During training, DNN weights are adjusted to improve network performance through repeated exposure to the labelled data-examples of a large dataset. During forward-inference, already trained networks are used to analyze new data-examples. |
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