Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

NTT technical review

Performance Comparison between Coherent Ising Machines and Quantum Annealer

Hiroki Takesue, Takahiro Inagaki, Kensuke Inaba, Toshimori Honjo

Published May 1, 2021• 3 citations
Publisher

Abstract

NTT has been developing a coherent Ising machine (CIM), which efficiently finds solutions to groundstate-search problems of the Ising model using a network of optical parametric oscillators.We introduce our recent experiment that compared the computational performance of CIMs (the one developed by NTT and another developed by Stanford University) with a quantum annealer, which solves the Isingmodel problems using superconducting devices.

DOI
10.53829/ntr202105fa2

Related papers

Synchronously-pumped OPO coherent Ising machine: benchmarking and prospects

Ryan Hamerly, Takahiro Inagaki, Peter L. McMahon +4 more

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.

AI and Optical Data Sciences•Feb 24, 2020•4 citations

The Future of Problem Solving: The Coherent Ising Machine Approach

Y. Yamamoto

Several new computing models tackle combinatorial optimization problems.These include adiabatic quantum computation, also known as quantum annealing, and the coherent Ising machine (CIM).The Physics and Informatics Laboratories at NTT Research launched a CIM initiative, and NTT Basic Research Laboratories built large-scale CIM prototypes.This article reviews the CIM approach, results to date, and future research agenda.

NTT technical review•Jan 1, 2022•0 citations

Experimental investigation of performance differences between coherent Ising machines and a quantum annealer

Ryan Hamerly, Takahiro Inagaki, Peter L. McMahon +20 more

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 measurement-feedback 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(–α_(DW)N^2)] relative to CIMs [exp(–α_(CIM)N)] 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.

•Jan 1, 2019•257 citations

A coherent Ising machine based on a network of 100,000 degenerate optical parametric oscillator pulses

Toshimori Honjo, Kensuke Inaba, Takahiro Inagaki +3 more

Coherent Ising machines (CIMs) are increasingly anticipated as a new computer that can solve combinatorial optimization problems. CIMs use degenerate optical parametric oscillator pulses as artificial spins. We constructed a 100,512-spin coherent Ising machine, the world's largest. Our CIM can find good solutions to maximum cut problems of 100,000-node graphs ~1000 times faster than simulated annealing. Furthermore, it provides both very good scores and a broad distribution when we operate it near the phase transition point.

IEEE International Conference on Nanotechnology (IEEE-NANO)•Jul 4, 2022•6 citations

Large-scale Coherent Ising Machine

Hiroki Takesue, Takahiro Inagaki, Kensuke Inaba +2 more

Today, world-wide efforts are being undertaken to explore non-von Neumann computation methods to overcome "the end of Moore's law". A promising approach is the coherent Ising machine (CIM), which utilizes a network of degenerate optical parametric oscillators for solving ground-state-search problems of the Ising model. Since its proposal in 2013 and the first proof-of-principle experiment in 2014, research on experimental CIMs has made rapid progress and now reached the level where we can find solutions to large-scale combinatorial optimization problems using a CIM system. In this paper, we review recent progress on CIMs, mainly focusing on the experimental side.

Journal of the Physical Society of Japan•Mar 1, 2019•31 citations

100,000-spin coherent Ising machine

Toshimori Honjo, Tomohiro Sonobe, Kensuke Inaba +9 more

Computers based on physical systems are increasingly anticipated to overcome the impending limitations on digital computer performance. One such computer is a coherent Ising machine (CIM) for solving combinatorial optimization problems. Here, we report a CIM with 100,512 degenerate optical parametric oscillator pulses working as the Ising spins. We show that the CIM delivers fine solutions to maximum cut problems of 100,000-node graphs drastically faster than standard simulated annealing. Moreover, the CIM, when operated near the phase transition point, provides some extremely good solutions and a very broad distribution. This characteristic will be useful for applications that require fast random sampling such as machine learning.

Science Advances•Sep 29, 2021•340 citations
NTT has been developing a coherent Ising machine (CIM), which efficiently finds solutions to groundstate-search problems of the Ising model using a network of optical parametric oscillators.We introduce our recent experiment that compared the computational performance of CIMs (the one developed by NTT and another developed by Stanford University) with a quantum annealer, which solves the Isingmodel problems using superconducting devices.
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.
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 measurement-feedback 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(–α_(DW)N^2)] relative to CIMs [exp(–α_(CIM)N)] 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.
Coherent Ising machines (CIMs) are increasingly anticipated as a new computer that can solve combinatorial optimization problems. CIMs use degenerate optical parametric oscillator pulses as artificial spins. We constructed a 100,512-spin coherent Ising machine, the world's largest. Our CIM can find good solutions to maximum cut problems of 100,000-node graphs ~1000 times faster than simulated annealing. Furthermore, it provides both very good scores and a broad distribution when we operate it near the phase transition point.
Today, world-wide efforts are being undertaken to explore non-von Neumann computation methods to overcome "the end of Moore's law". A promising approach is the coherent Ising machine (CIM), which utilizes a network of degenerate optical parametric oscillators for solving ground-state-search problems of the Ising model. Since its proposal in 2013 and the first proof-of-principle experiment in 2014, research on experimental CIMs has made rapid progress and now reached the level where we can find solutions to large-scale combinatorial optimization problems using a CIM system. In this paper, we review recent progress on CIMs, mainly focusing on the experimental side.
Computers based on physical systems are increasingly anticipated to overcome the impending limitations on digital computer performance. One such computer is a coherent Ising machine (CIM) for solving combinatorial optimization problems. Here, we report a CIM with 100,512 degenerate optical parametric oscillator pulses working as the Ising spins. We show that the CIM delivers fine solutions to maximum cut problems of 100,000-node graphs drastically faster than standard simulated annealing. Moreover, the CIM, when operated near the phase transition point, provides some extremely good solutions and a very broad distribution. This characteristic will be useful for applications that require fast random sampling such as machine learning.