Ising Machine Search Architecture for Faster Ground State Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems solving combinatorial optimization problems, such as the Ising problem, face inefficiencies in information transmission and processing time between host units and Ising machines, hindering the speed and accuracy of finding ground states in Ising models.
Innovation Solution
An information processing system comprising an Ising machine and a host unit, where the Ising machine performs a search process by alternately updating main and auxiliary variables, and the host unit transmits initial values and receives optimized spin values, utilizing a simulated bifurcation method to efficiently solve Ising problems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the host unit transmits the J matrix and h vector to the Ising machine and receives optimized spin values, then the combinatorial optimization problem can be solved, but the information transmission and processing time between host unit and Ising machine increases
Solution Approach 1:
The patent segments the optimization variables into two distinct groups: main variables (x) and auxiliary variables (y). This segmentation allows the system to update variables in alternating steps, where the Ising machine updates main variables while the host unit updates auxiliary variables. By dividing the variable set and updating them separately in alternating fashion, the system reduces the amount of data that needs to be transmitted between host unit and Ising machine at each step, thereby reducing communication overhead and processing time while maintaining solution accuracy.
Solution Approach 2:
The patent performs preliminary actions by having the host unit update auxiliary variables before the Ising machine updates main variables in each iteration. The auxiliary variables are updated based on the current main variable values, and these updated auxiliary variables are then used in the subsequent main variable update step. This preliminary updating of auxiliary variables allows the system to prepare optimization information in advance, reducing the computational burden during the main optimization step and overall processing time.
2Measurement precision
If the Ising machine performs search process for ground state, then optimization accuracy is improved, but the time from start of calculation to output of solution increases
Solution Approach 1:
The patent implements continuous useful action through alternating updates of main and auxiliary variables without idle waiting periods. While the Ising machine is updating main variables, the host unit simultaneously prepares the next set of auxiliary variable updates. This continuous alternating process eliminates idle time where one component would be waiting for the other, maintaining high utilization of both the Ising machine and host unit throughout the optimization process, thereby improving calculation speed without sacrificing accuracy.
Solution Approach 2:
By segmenting the variable updates into alternating main variable updates (performed by Ising machine) and auxiliary variable updates (performed by host unit), the patent enables parallel processing of different variable types. This segmentation allows the system to perform useful work continuously on both fronts simultaneously rather than sequentially, improving overall calculation speed while maintaining the rigorous ground state search process.
Data Source
AI summary
According to an embodiment, an information processing system solves a combinatorial optimization problem. The information processing system includes an Ising machine and a host unit. The Ising machine is hardware configured to perform a search process for searching for the ground state of an Ising model that represents the combinatorial optimization problem. The host unit is hardware connected to the Ising machine via an interface and configured to control the Ising machine. In the search process, for each of a plurality of Ising spins, the Ising machine alternately repeats an auxiliary variable update process for updating an auxiliary variable by a main variable and a main variable update process for updating the main variable by the auxiliary variable multiple times. Prior to the search process, the host unit transmits, to the Ising machine, an initial value of the auxiliary variable corresponding to each of the plurality of Ising spins.