Search

© COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only. Storing, proceßing, and learning from data is a central task in both industrial practice and modern science. Recent advances in modern statistical learning, particularly Deep Neural Networks (DNNs), have given record breaking performance on tasks in game playing,1, 2 natural language proceßing,3 computer vision,4 computational biology,5, 6 and many others. The rapid growth of the field has been driven by an increase in the amount of public datasets,7 improvements to algorithms,8 and a substantial growth in computing power.9 In order to perform well on these tasks networks have had to grow in size, learning more complicated statistical features. The training and deployment of these large neural networks has spurred the creation of many neural network accelerators to aid in the computation of these networks.10-12 Existing general purpose computing devices such as CPUs and GPUs are limited both by thermal dißipation per unit area and yield aßociated with large chips.13, 14 The design of Application Specific Integrated circuits (ASICs) has aided in decreasing the energy consumption per workload substantially by limiting the supported operations on chip. An example of this is the first generation tensor proceßing unit (TPU)15 which is able to perform the inference of large convolutional neural networks in datacenter in <10ms with an idle power of 28W and an workload power of 40W. It may seen counterintuitive then that the limiting factor for the implementation of DNNs is not computation, but rather the energy and bandwidth aßociated with reading and writing data from memory as well as the energy cost of moving data inside of the ASIC.15, 16 Several emerging technologies, such as in-memory computing,17 memristive croßbar arrays18 promise increased performance, but these emerging architectures suffer from calibration ißues and limited accuracy.19 Photonics as a field has had tremendous succeß

© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement Conventional computing architectures have no known efficient algorithms for combinatorial optimization tasks such as the Ising problem, which requires finding the ground state spin configuration of an arbitrary Ising graph. Physical Ising machines have recently been developed as an alternative to conventional exact and heuristic solvers; however, these machines typically suffer from decreased ground state convergence probability or universality for high edge-density graphs or arbitrary graph weights, respectively. We experimentally demonstrate a proof-of-principle integrated nanophotonic recurrent Ising sampler (INPRIS), using a hybrid scheme combining electronics and silicon-on-insulator photonics, that is capable of converging to the ground state of various four-spin graphs with high probability. The INPRIS results indicate that noise may be used as a resource to speed up the ground state search and to explore larger regions of the phase space, thus allowing one to probe noise-dependent physical observables. Since the recurrent photonic transformation that our machine imparts is a fixed function of the graph problem and therefore compatible with optoelectronic architectures that support GHz clock rates (such as passive or non-volatile photonic circuits that do not require reprogramming at each iteration), this work suggests the potential for future systems that could achieve orders-of-magnitude speedups in exploring the solution space of combinatorially hard problems.

© 2019 by the Authors. Physical annealing systems provide heuristic approaches to solving combinatorial optimization problems. Here, we benchmark two types of annealing machines-a quantum annealer built by D-Wave Systems and measurementfeedback coherent Ising machines (CIMs) based on optical parametric oscillators-on two problem classes, the Sherrington-Kirkpatrick (SK) model and MAX-CUT. The D-Wave quantum annealer outperforms the CIMs on MAX-CUT on cubic graphs. On denser problems, however, we observe an exponential penalty for the quantum annealer [exp(-aDWN2)] relative to CIMs [exp(-aCIMN)] for fixed anneal times, both on the SK model and on 50% edge density MAX-CUT. This leads to a several orders of magnitude time-to-solution difference for instances with over 50 vertices. An optimal-annealing time analysis is also consistent with a substantial projected performance difference. The difference in performance between the sparsely connected D-Wave machine and the fully-connected CIMs provides strong experimental support for efforts to increase the connectivity of quantum annealers.

We propose a digital incoherent optical neural network architecture using the passive data routing and copying capabilities of optics for artificial neural network acceleration. We demonstrate a proof-of-concept experiment and analyze optimal use cases.

The coherent Ising machine (CIM) is a network of optical parametric oscillators (OPOs) that solves for the ground state of Ising problems through OPO bifurcation dynamics. Here, we present experimental results comparing the performance of the CIM to quantum annealers (QAs) on two claßes of NP-hard optimization problems: ground state calculation of the Sherrington-Kirkpatrick (SK) model and MAX-CUT. While the two machines perform comparably on sparsely-connected problems such as cubic MAX-CUT, on problems with dense connectivity, the QA shows an exponential performance penalty relative to CIMs. We attribute this to the embedding overhead required to map dense problems onto the sparse hardware architecture of the QA, a problem that can be overcome in photonic architectures such as the CIM., United States. National Aeronautics and Space Administration. Academic Mission Services (Grant NNA16BD14C), United States. Army Research Office (Grants W911NF-18-2-0048 and SRC-NSF E2CDA)