Comparison of annealing methods for photonics Ising machines

Toon Sevenants; Guy Verschaffelt; Guy Van Der Sande

Comparison of annealing methods for photonics Ising machines

Toon Sevenants, Guy Verschaffelt, Guy Van Der Sande

Research output: Unpublished contribution to conference › Poster

Abstract

Ising machines (IM) are a natural computing platform that could solve complex optimization problems more efficiently than traditional
digital computers. Unfortunately, for specific tasks, the relatively low success rate of finding the optimal solution remains an issue.
Annealing schemes are being investigated as a tool to navigate around these local minima and increase the overall success rate of IM.
Our research investigates the influence of linear-annealing schemes as well as adiabatic annealing schemes on the success rate of the
IM. Our results indicate that both annealing schemes can be useful to increase the success rate of some benchmark problems.

Original language	English
Number of pages	1
Publication status	Published - 23 Nov 2023
Event	27th IEEE Photonics Benelux Symposium 2023 - De Abdij van Drongen, Drongen, Belgium Duration: 23 Nov 2023 → 24 Nov 2023 https://www.ieee.be/node/1040

Conference

Conference	27th IEEE Photonics Benelux Symposium 2023
Country/Territory	Belgium
City	Drongen
Period	23/11/23 → 24/11/23
Internet address	https://www.ieee.be/node/1040

Keywords

Ising machines
annealing
Adiabatic annealing
photonic computing

Handle.net

Persistent link

FWOEOS20: Photonic Ising Machines
Van Der Sande, G. (Administrative Promotor) & Verschaffelt, G. (CoI (Co-Promotor))
1/01/22 → 31/12/25
Project: Fundamental

Cite this

@conference{98853a4661f34674a8ae782046c66b30,

title = "Comparison of annealing methods for photonics Ising machines",

abstract = "Ising machines (IM) are a natural computing platform that could solve complex optimization problems more efficiently than traditionaldigital computers. Unfortunately, for specific tasks, the relatively low success rate of finding the optimal solution remains an issue.Annealing schemes are being investigated as a tool to navigate around these local minima and increase the overall success rate of IM.Our research investigates the influence of linear-annealing schemes as well as adiabatic annealing schemes on the success rate of theIM. Our results indicate that both annealing schemes can be useful to increase the success rate of some benchmark problems.",

keywords = "Ising machines, annealing, Adiabatic annealing, photonic computing",

author = "Toon Sevenants and Guy Verschaffelt and {Van Der Sande}, Guy",

year = "2023",

month = nov,

day = "23",

language = "English",

note = "27th IEEE Photonics Benelux Symposium 2023 ; Conference date: 23-11-2023 Through 24-11-2023",

url = "https://www.ieee.be/node/1040",

}

TY - CONF

T1 - Comparison of annealing methods for photonics Ising machines

AU - Sevenants, Toon

AU - Verschaffelt, Guy

AU - Van Der Sande, Guy

PY - 2023/11/23

Y1 - 2023/11/23

N2 - Ising machines (IM) are a natural computing platform that could solve complex optimization problems more efficiently than traditionaldigital computers. Unfortunately, for specific tasks, the relatively low success rate of finding the optimal solution remains an issue.Annealing schemes are being investigated as a tool to navigate around these local minima and increase the overall success rate of IM.Our research investigates the influence of linear-annealing schemes as well as adiabatic annealing schemes on the success rate of theIM. Our results indicate that both annealing schemes can be useful to increase the success rate of some benchmark problems.

AB - Ising machines (IM) are a natural computing platform that could solve complex optimization problems more efficiently than traditionaldigital computers. Unfortunately, for specific tasks, the relatively low success rate of finding the optimal solution remains an issue.Annealing schemes are being investigated as a tool to navigate around these local minima and increase the overall success rate of IM.Our research investigates the influence of linear-annealing schemes as well as adiabatic annealing schemes on the success rate of theIM. Our results indicate that both annealing schemes can be useful to increase the success rate of some benchmark problems.

KW - Ising machines

KW - annealing

KW - Adiabatic annealing

KW - photonic computing

M3 - Poster

T2 - 27th IEEE Photonics Benelux Symposium 2023

Y2 - 23 November 2023 through 24 November 2023

ER -

Comparison of annealing methods for photonics Ising machines

Abstract

Conference

Keywords

Handle.net

Fingerprint

Projects

FWOEOS20: Photonic Ising Machines

Cite this