Coherent Ising Matrix Parameter Calculation for Faster Solving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for selecting matrix parameters for a coherent Ising machine are manually intensive, slow, and have a low success rate.
Innovation Solution
Automatically calculate matrix parameters using formulas based on an input matrix, specifically through formulas such as \( s = \sum_{i,j} J_{ij} \), \( \alpha = 0.7s + 1 \), and \( eta = 0.7ss + 1 \), to derive a target matrix \( Q = \alpha I + eta J \) for the coherent Ising machine.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual selection of matrix parameters is used, then engineers can adjust parameters, but the feedback speed is slow and success rate is low
Solution Approach 1:
The system performs self-service by automatically calculating optimal matrix parameters through the formula Q = αI + βJ, where α and β are derived from the input matrix J. This eliminates the need for manual parameter adjustment by engineers and enables the system to self-determine the best parameters for solving Ising problems.
Solution Approach 2:
The patent applies parameter changes by transforming the input matrix J into optimized parameters α and β through mathematical relationships (α = 0.7s + 1, β = 0.7ss + 1 where s is the target parameter calculated from J). This systematic parameter transformation replaces manual trial-and-adjustment with deterministic calculation, improving both speed and success rate.
2Adaptability or versatility
If manual parameter adjustment is used, then engineers can try different combinations, but the method is prone to taking money from one pocket to pay for another and has low success rate
Solution Approach 1:
The patent systematically changes parameters by establishing mathematical relationships between the input matrix J and the optimized parameters α and β. Instead of arbitrary manual adjustments, the system uses deterministic formulas (α = 0.7s + 1, β = 0.7ss + 1) that guarantee optimal parameter combinations for any given input matrix, thereby improving reliability while maintaining adaptability.
Solution Approach 2:
The system implements feedback by calculating the target parameter s from the input matrix J (s = ΣJij/Σ|Jij|) and using this feedback to determine the optimal parameters α and β. This closed-loop approach ensures that parameter selection is based on actual input characteristics rather than manual guessing, improving solution success rate.
3Productivity
If automated calculation is used, then the process is faster, but requires mathematical formulas and calculations
Solution Approach 1:
The patent simplifies calculation complexity by establishing direct linear relationships between input matrix properties and output parameters. The formulas α = 0.7s + 1 and β = 0.7ss + 1 are simple polynomial transformations that can be efficiently computed, achieving high productivity without requiring complex algorithms or iterative optimization processes.
Data Source
AI summary
A method and device for acquiring matrix parameters of coherent Ising machine, and an electronic device are provided. The method includes: receiving an input matrix of the coherent Ising machine; calculating, based on the input matrix, a target parameter of the coherent Ising machine; calculating, based on the target parameter, a first matrix parameter and a second matrix parameter of the coherent Ising machine; calculating, based on the input matrix, the first matrix parameter and the second matrix parameter, a target matrix of the coherent Ising machine; and inputting the target matrix into the coherent Ising machine. The method automatically calculates two key and appropriate matrix parameters through formulas, and substitutes these parameters into preset matrix formulas to obtain the target matrix. It is no longer necessary for engineers to try different parameters and give calculation results, which greatly improves a solving success rate of the coherent Ising machine.