Superconducting neural network for solving a combinatorial optimization problem

Takeshi Onomi, Yusuke Maenami, Koji Nakajima

Research output: Contribution to journalArticlepeer-review

19 Citations (Scopus)

Abstract

We propose a neural network using coupled -SQUIDs to solve the N-Queens problem, a combinatorial optimization problem. The N-Queen problem consists of placing N queens on an N × N chess board such that none of the queens are able to capture any other using standard chess moves for a queen. We run a numerical simulation to show that a network consisting of a combination of coupled-SQUIDs can arrive at the solution. However, conditions of the network may be trapped in incorrect answers due to the existence of local minima on the energy function of the network. The Josephson voltage oscillation effect is effective for escaping from such conditions due to the existence of local minima. We investigate network dynamics and discuss the performance of the network on the basis of the parameters of the Nb integration circuit.

Original languageEnglish
Article number5672805
Pages (from-to)701-704
Number of pages4
JournalIEEE Transactions on Applied Superconductivity
Volume21
Issue number3 PART 1
DOIs
Publication statusPublished - 2011 Jun

Keywords

  • Combinatorial optimization
  • N-Queens problem
  • Neural networks
  • Superconducting integrated circuits

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Superconducting neural network for solving a combinatorial optimization problem'. Together they form a unique fingerprint.

Cite this