Binary Model Partitioning for Ising Computer Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Combinatorial optimization problems with a large number of variables exceed the capacity of existing Ising computers, requiring methods to reduce variables and partition problems into sub-problems, but existing methods lack effective analysis for constraint conditions, especially for binary models.
Innovation Solution
An optimization apparatus and method that partitions binary combinatorial optimization problems into sub-models by generating binary sub-models where incompatibility constraint conditions form connected components, allowing for efficient processing on Ising computers, and handles constraint conditions by prioritizing variables that cannot simultaneously take a specific value.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If a combinatorial optimization problem with a large number of variables is solved using an Ising computer, then the problem can be solved using quantum annealing or simulated annealing, but the problem exceeds the fixed variable capacity of the Ising computer
Solution Approach 1:
The patent partitions the combinatorial optimization problem into multiple sub-problems, each with fewer variables that fit within the Ising computer's capacity. The partitioning is performed by analyzing constraint conditions and dividing variables into groups that can be independently processed, then combining the results to obtain the solution to the original problem.
2Productivity
If the combinatorial optimization problem is partitioned into sub-problems by reducing variables, then the problem can be solved within Ising computer capacity, but existing methods lack effective analysis for constraint conditions
Solution Approach 1:
The patent performs preliminary analysis of constraint conditions before partitioning the problem. By identifying and analyzing constraint conditions in advance, the method determines an effective partitioning strategy that reduces subsequent processing complexity and improves solution efficiency.
Solution Approach 2:
The patent introduces constraint condition analysis as an intermediary step between problem formulation and partitioning. This intermediary process extracts key relationships from constraint conditions and uses them to guide the partitioning, making the overall process more systematic and efficient.
3Ease of manufacture
If variables are partitioned without considering constraint conditions, then the partitioning process is simpler, but the solution accuracy and effectiveness are reduced
Solution Approach 1:
The patent applies different partitioning strategies to different regions or groups of variables based on their constraint relationships. By analyzing constraint conditions locally for each variable group and applying targeted partitioning methods, the approach maintains high solution accuracy while managing partitioning complexity.
Data Source
AI summary
An optimization apparatus provided with at least one memory configured to store instructions; and at least one processor configured to execute the instructions to partition a binary model representing a combinatorial optimization problem to generate binary sub-models. The at least one processor is configured to generate the binary sub-models such that graphs indicating incompatibility constraint conditions of variables in the binary sub-models form connected components, and the incompatibility constraint conditions are constraint conditions indicating that two values that can be taken by the variables cannot simultaneously be one specific value of the two values.