Generalizing Pauli Rotation Synthesis in Quantum Circuits

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for synthesizing quantum circuits, such as Graysynth and Unitary Coupled Cluster Ansatz, are limited in their ability to optimize Pauli rotations and require complex pre-processing to restrict rotations to the Z-axis, leading to inefficiencies and high computational time.

Innovation Solution

A generalized method that transforms quantum circuits into tables to reveal entanglement and rotation patterns, allowing for recursive application of reduced quantum gates to synthesize circuits with any type of Pauli rotations, optimizing CNOT count and automation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If Graysynth algorithm is used to synthesize quantum circuits, then the number of CNOT gates is reduced, but the method can be applied only to circuits with rotational quantum gates along Z-axis

Engineering Contradiction:
ImproveCNOT gate countVSAvoidapplicability to different rotation types
Core Design Contradiction:
Quantity of substanceVSAdaptability or versatility

Solution Approach 1:

The patent generalizes the Graysynth algorithm to handle any Pauli rotation type (X, Y, Z axes) by introducing a classification mechanism for pivot rows based on their rotation axis. The algorithm maintains the optimized CNOT synthesis capability while becoming universally applicable to all single-qubit rotation types through systematic row classification and targeted gate insertion.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent changes the parameter representation from restricted Z-axis rotations to general Pauli rotations by classifying pivot rows into different categories (X-type, Y-type, Z-type). This parameter expansion allows the algorithm to adapt its gate insertion strategy based on the specific rotation type, maintaining optimization while increasing versatility.

Inventive Principle:
Principle #35Parameter changes

2Quantity of substance

If Unitary Coupled Cluster Ansatz method is used to compile circuits, then depth and number of CNOT gates are reduced, but exponential number of steps and time are required for pre-processing

Engineering Contradiction:
ImproveCNOT gate count and circuit depthVSAvoidpre-processing time
Core Design Contradiction:
Quantity of substanceVSLoss of time

Solution Approach 1:

The patent extracts and eliminates the time-consuming pre-processing step of modifying circuits to obtain only Z-axis rotations. By directly handling general Pauli rotations through classification and targeted synthesis, the method removes the intermediate transformation step while maintaining the optimized circuit output.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent performs preliminary classification of pivot rows by their rotation type before synthesis. This preliminary organization allows the algorithm to directly apply appropriate synthesis strategies without requiring subsequent complex transformations, thereby reducing overall processing time while maintaining optimization quality.

Inventive Principle:
Principle #10Preliminary action

3Manufacturing precision

If circuit transformation is performed to obtain only rotational quantum gates along Z-axis, then optimized compiled circuits are obtained, but the transformation is complex and time consuming

Engineering Contradiction:
Improvecircuit optimization qualityVSAvoidtransformation complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

Instead of transforming all rotations to Z-axis as the conventional approach, the patent inverts the strategy by directly synthesizing circuits for the given rotation types through classification. This inversion eliminates the transformation step while achieving comparable or better optimization by working with the original rotation types.

Inventive Principle:
Principle #13The other way round (Inversion)

Solution Approach 2:

The patent applies different synthesis strategies locally based on the rotation type of each pivot row. Rather than forcing a uniform Z-axis transformation across the entire circuit, the algorithm selects appropriate synthesis methods for X-type, Y-type, and Z-type rows individually, reducing overall complexity while maintaining local optimization quality.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS12118432B2Method for synthesizing product of Pauli rotations in a quantum circuit and process for synthesizing quantum circuits for Trotter-Suzuki n-order expansion
Publication Date: 2024.10.15 BULL SA
  • US12118432B2 patent drawing
  • US12118432B2 patent drawing
  • US12118432B2 patent drawing

AI summary

A method for generalizing an algorithm configured to synthesize a diagonal product of Pauli rotations to synthesize a product of Pauli rotations comprising X, Y and Z rotations, the method comprising:Providing a table of p number of rows and m number of columns, where p is a number of qubits and m a number of rotations in the quantum circuit, and where the table comprises X, Y, Z or I entry corresponding to the respective rotations of the qbits;Determining a pivot row,and recursively, until all rotations of the product of Pauli rotations are 1-qubit rotations:Determine a target row,Conjugate the target row with the pivot row by insertion of predetermined quantum gates on the qubits corresponding to the target row and/or pivot row by calling, at each recursive call, entries of the same type of the pivot row and by always calling first the identity entry.