Qubit Allocation via Graph Isomorphism for Faster Circuit Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional qubit allocation techniques require extensive calculation time due to the need to match quantum circuits with a large number of templates, which is inefficient and time-consuming.
Innovation Solution
A qubit allocation program and device that convert quantum circuits into canonical forms, remove non-essential gates, and perform graph isomorphism to match with pre-calculated templates, reducing the number of gates and templates needed for efficient qubit allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional qubit allocation techniques use a large number of templates for pattern matching, then qubit allocation accuracy is improved, but calculation time increases significantly
Solution Approach 1:
The patent extracts and removes non-essential gates (single-qubit gates, measurement gates, reset gates) from the quantum circuit graph, retaining only two-qubit gates that are critical for qubit allocation. This extraction reduces the graph size and complexity while preserving the essential allocation information, thereby reducing calculation time without sacrificing allocation accuracy
Solution Approach 2:
The patent segments the quantum circuit analysis into two parts: (1) removing non-essential gates to create a simplified graph for fast template matching, and (2) using pre-calculated template results for qubit allocation. This segmentation allows the system to achieve both speed and accuracy by handling different aspects of the allocation problem separately
2Productivity
If the number of gates in the quantum circuit graph is reduced, then calculation time is reduced, but the ability to maintain allocation accuracy may be compromised
Solution Approach 1:
The patent applies local quality by treating different types of gates differently: two-qubit gates are retained in the graph because they locally determine qubit connectivity and allocation constraints, while single-qubit gates are removed as they do not affect allocation. This selective retention based on local gate properties maintains accuracy where it matters while improving speed overall
Solution Approach 2:
The patent performs preliminary action by pre-calculating qubit allocation for template circuits and storing the results. When a new quantum circuit matches a template pattern, the pre-calculated allocation is directly applied, eliminating the need for recalculation and ensuring accuracy is maintained through proven template solutions
Data Source
AI summary
A non-transitory computer-readable recording medium stores therein a qubit allocation program that causes a computer to execute a process including converting a quantum circuit serving as an operation target into a first graph indicating a connection relation of each gate included in the quantum circuit, first creating a second graph from which a gate other than a two-quantum gate included in the first graph is removed, based on the first graph, specifying a graph isomorphic to the second graph from a third graph created based on each of a plurality of the quantum circuits, and second creating circuit information corresponding to the quantum circuit serving as the operation target, to which a qubit is allocated based on the quantum circuits corresponding to the specified third graph.