Experimental Demonstration of a Reconfigurable Coupled Oscillator Platform to Solve the Max-Cut Problem

In this work, we experimentally demonstrate an integrated circuit (IC) of 30 relaxation oscillators with reconfigurable capacitive coupling to solve the NP-Hard maximum cut (Max-Cut) problem. We show that under the influence of an external second-harmonic injection signal, the oscillator phases exhi...

Full description

Bibliographic Details
Main Authors: Mohammad Khairul Bashar, Antik Mallick, Daniel S. Truesdell, Benton H. Calhoun, Siddharth Joshi, Nikhil Shukla
Format: Article
Language:English
Published: IEEE 2020-01-01
Series:IEEE Journal on Exploratory Solid-State Computational Devices and Circuits
Subjects:
Online Access:https://ieeexplore.ieee.org/document/9204635/
_version_ 1818370113326284800
author Mohammad Khairul Bashar
Antik Mallick
Daniel S. Truesdell
Benton H. Calhoun
Siddharth Joshi
Nikhil Shukla
author_facet Mohammad Khairul Bashar
Antik Mallick
Daniel S. Truesdell
Benton H. Calhoun
Siddharth Joshi
Nikhil Shukla
author_sort Mohammad Khairul Bashar
collection DOAJ
description In this work, we experimentally demonstrate an integrated circuit (IC) of 30 relaxation oscillators with reconfigurable capacitive coupling to solve the NP-Hard maximum cut (Max-Cut) problem. We show that under the influence of an external second-harmonic injection signal, the oscillator phases exhibit a bipartition that can be used to calculate a high-quality approximate Max-Cut solution. Leveraging the all-to-all reconfigurable coupling architecture, we experimentally evaluate the computational properties of the oscillators using randomly generated graph instances of varying size and edge density (η). Furthermore, comparing the Max-Cut solutions with the optimal values, we show that the oscillators (after simple postprocessing) produce a Max-Cut that is within 99% of the optimal value in 28 of the 36 measured graphs; importantly, the oscillators are particularly effective in dense graphs with the Max-Cut being optimal in seven out of nine measured graphs with η = 0.8. Our work marks a step toward creating an efficient, room-temperature-compatible non-Boolean hardware-based solver for hard combinatorial optimization problems.
first_indexed 2024-12-13T23:34:34Z
format Article
id doaj.art-ebaf7f5bbf594e1686b3e7bdf38bd6a5
institution Directory Open Access Journal
issn 2329-9231
language English
last_indexed 2024-12-13T23:34:34Z
publishDate 2020-01-01
publisher IEEE
record_format Article
series IEEE Journal on Exploratory Solid-State Computational Devices and Circuits
spelling doaj.art-ebaf7f5bbf594e1686b3e7bdf38bd6a52022-12-21T23:27:20ZengIEEEIEEE Journal on Exploratory Solid-State Computational Devices and Circuits2329-92312020-01-016211612110.1109/JXCDC.2020.30259949204635Experimental Demonstration of a Reconfigurable Coupled Oscillator Platform to Solve the Max-Cut ProblemMohammad Khairul Bashar0https://orcid.org/0000-0002-3668-0576Antik Mallick1https://orcid.org/0000-0002-5697-5742Daniel S. Truesdell2https://orcid.org/0000-0002-5723-0398Benton H. Calhoun3Siddharth Joshi4Nikhil Shukla5https://orcid.org/0000-0002-8899-5190Department of Electrical & Computer Engineering, University of Virginia, Charlottesville, VA, USADepartment of Electrical & Computer Engineering, University of Virginia, Charlottesville, VA, USADepartment of Electrical & Computer Engineering, University of Virginia, Charlottesville, VA, USADepartment of Electrical & Computer Engineering, University of Virginia, Charlottesville, VA, USADepartment of Computer Science and Engineering, University of Notre Dame, Notre Dame, IN, USADepartment of Electrical & Computer Engineering, University of Virginia, Charlottesville, VA, USAIn this work, we experimentally demonstrate an integrated circuit (IC) of 30 relaxation oscillators with reconfigurable capacitive coupling to solve the NP-Hard maximum cut (Max-Cut) problem. We show that under the influence of an external second-harmonic injection signal, the oscillator phases exhibit a bipartition that can be used to calculate a high-quality approximate Max-Cut solution. Leveraging the all-to-all reconfigurable coupling architecture, we experimentally evaluate the computational properties of the oscillators using randomly generated graph instances of varying size and edge density (η). Furthermore, comparing the Max-Cut solutions with the optimal values, we show that the oscillators (after simple postprocessing) produce a Max-Cut that is within 99% of the optimal value in 28 of the 36 measured graphs; importantly, the oscillators are particularly effective in dense graphs with the Max-Cut being optimal in seven out of nine measured graphs with η = 0.8. Our work marks a step toward creating an efficient, room-temperature-compatible non-Boolean hardware-based solver for hard combinatorial optimization problems.https://ieeexplore.ieee.org/document/9204635/Analogcoupled oscillatorsintegrated circuit (IC)Ising machinesmaximum cut (Max-Cut)
spellingShingle Mohammad Khairul Bashar
Antik Mallick
Daniel S. Truesdell
Benton H. Calhoun
Siddharth Joshi
Nikhil Shukla
Experimental Demonstration of a Reconfigurable Coupled Oscillator Platform to Solve the Max-Cut Problem
IEEE Journal on Exploratory Solid-State Computational Devices and Circuits
Analog
coupled oscillators
integrated circuit (IC)
Ising machines
maximum cut (Max-Cut)
title Experimental Demonstration of a Reconfigurable Coupled Oscillator Platform to Solve the Max-Cut Problem
title_full Experimental Demonstration of a Reconfigurable Coupled Oscillator Platform to Solve the Max-Cut Problem
title_fullStr Experimental Demonstration of a Reconfigurable Coupled Oscillator Platform to Solve the Max-Cut Problem
title_full_unstemmed Experimental Demonstration of a Reconfigurable Coupled Oscillator Platform to Solve the Max-Cut Problem
title_short Experimental Demonstration of a Reconfigurable Coupled Oscillator Platform to Solve the Max-Cut Problem
title_sort experimental demonstration of a reconfigurable coupled oscillator platform to solve the max cut problem
topic Analog
coupled oscillators
integrated circuit (IC)
Ising machines
maximum cut (Max-Cut)
url https://ieeexplore.ieee.org/document/9204635/
work_keys_str_mv AT mohammadkhairulbashar experimentaldemonstrationofareconfigurablecoupledoscillatorplatformtosolvethemaxcutproblem
AT antikmallick experimentaldemonstrationofareconfigurablecoupledoscillatorplatformtosolvethemaxcutproblem
AT danielstruesdell experimentaldemonstrationofareconfigurablecoupledoscillatorplatformtosolvethemaxcutproblem
AT bentonhcalhoun experimentaldemonstrationofareconfigurablecoupledoscillatorplatformtosolvethemaxcutproblem
AT siddharthjoshi experimentaldemonstrationofareconfigurablecoupledoscillatorplatformtosolvethemaxcutproblem
AT nikhilshukla experimentaldemonstrationofareconfigurablecoupledoscillatorplatformtosolvethemaxcutproblem